Back to Table of Contents
The numbers after the model name (e.g. "CDR-102 (4x2/512K)") refer to the read and write speeds of the unit and the size of the write buffer. "4x2" would be a double-speed writer that's also a quad-speed reader. If it just says "?x2", the write speed is double-speed and the read speed isn't known (but presumably is at least 2x). Buffer sizes written with a '+', e.g. "2MB+", indicate that the buffer can be expanded further.
If the recorder can write to CD-RW media, the specification will include a third value, e.g. "6x4x2/1MB" would be a drive that reads at 6x, writes to CD-R media at 4x, and writes to CD-RW media at 2x.
Some manufacturers present the speed ratings in a different order, often write/rewrite/read. Some drives that support reading of DVD-ROM will be written write/rewrite/read/dvd-read. There is no standard approach.
Many units are repackaged versions of other manufacturer's devices, sometimes with slight changes in the firmware. Value-added retailers have been known to switch to a different manufacturer's drive without notice, so don't assume that everything here is accurate.
The interface is listed for each drive. "SCSI" means any form of SCSI (SCSI-2, SCSI-3, wide, narrow, ultra, etc). IDE means any ATAPI interface (e.g. Ultra-DMA/33). USB (1.x or 2.x), parallel-port, FireWire, and PCMCIA refer to interface styles for external drives (which are usually just internal ATAPI devices placed inside an enclosure with a power supply and an ATAPI converter).
Many of the models listed have been discontinued in favor of newer models, and some of them have yet to be released, so you will probably not be able to find all of the models listed here for sale.
If you're new to SCSI, take a look at the comp.periphs.scsi FAQ, http://www.faqs.org/faqs/scsi-faq. It covers both novice and advanced questions. If you want specs, try http://www.t10.org/.
A wealth of information on Enhanced IDE and other storage technologies is available from http://thef-nym.sci.kun.nl/~pieterh/storage.html and http://www.faqs.org/faqs/pc-hardware-faq/enhanced-IDE/part1/. If you want to debate the merits of SCSI vs EIDE, please read http://thef-nym.sci.kun.nl/~pieterh/eide-vs-scsi.html.
Some brief notes: ATA (AT Attachment Interface) is the official name for IDE (Integrated Drive Electronics) interfaces. ATAPI is the ATA Packet Interface, commonly used for controlling CD-ROM and tape devices. The ATA-2 changes grew out of vendor-specific "Enhanced IDE" implementations. There have been subsequent enhancements (ATA-3, ATA/ATAPI-4, etc).
There are no absolutely perfect recorders, but some drives are better than others. The best are listed below, and the risky propositions are identified with "CAVEAT EMPTOR" warnings in the individual sections. In cases where a unit is built by company A and repackaged by company B, the warnings are listed with the original manufacturer (company A).
http://www.storagereview.com/ has links to reviews of storage devices, including CD recorders. http://www.cdrlabs.com/ has reviews of both hardware and software products.
This section used to list specific models that were highly regarded. The manufacturers are coming out with new models so quickly that it's impossible to keep up in an FAQ that is updated monthly. The list of older models is below. See section (8-4) for a list of sites that carry news articles about newly released products.
As of late 2001, the most commonly recommended manufacturers (in no particular order) were:
If you have specific needs, you should verify with the manufacturer that the drive will do what you want. All computer-based recorders can create audio CDs and CD-ROMs, but some have additional features and some are just plain better at it. If you want a drive that works well with a specific piece of software, e.g. CloneCD (6-1-49), then you should check the web page for that software to see which drives they recommend.
Some older models that would be worth having are listed below. As of October 2001 this list is no longer being updated.
The model numbers are important! Sometimes the older or newer models from the same manufacturer aren't as good. The units listed were considered independently from the software that they were bundled with, and it may be necessary to buy additional software to get the full value from the drive.
External drives were traditionally preferred to internal drives because of heat problems, but this is only a minor concern for current models. External models do have the advantage that they can be moved between machines, and even between platforms. Most if not all SCSI models will work on both Macs and PCs, as should USB recorders.
I'm not currently listing stand-alone recorders like the "CD Blaster" or "CD Dupe-It", which are boxes with a CPU, CD-R, and hard drive that can duplicate CDs without tying up a full machine. Most of these low-end CD production boxes are off-the-shelf hardware and software packaged into a single unit, so listing them separately doesn't make much sense. Besides, they're not of much interest to the average user. Interested users can find some relevant URLs in (5-19).
CDR-100 (4x4/512K;SCSI)[ Yamaha departed the optical storage market in February 2003. ]
CDR-400 (6x4/2MB;SCSI; 'c' is caddy, 't' is tray, 'x' is external)
CRW-4416 (16x4x4/2MB; 'S'=SCSI, 'E'=IDE)
CRW-8824 (24x8x8/4MB; 'S'=SCSI, 'E'=IDE, 'F'=Firewire, 'X'=external)
CRW-2100 (40x16x10/8MB; 'S'=SCSI-int, 'SX'=SCSI-ext, 'E'=IDE, 'IX'=Firewire)
CRW-2200E (40x20x10/8MB; 'SX'=SCSI-ext, 'E'=IDE, 'IX'=FireWire, 'UX'=USB)
CRW-3200 (40x24x10/8MB; 'SX'=SCSI-ext, 'E'=IDE, 'IX'=FireWire, 'UX'=USB)
It has been reported that the CDR-102 is the same mechanism as the CDR-100, but with the 4x writing feature disabled. There is no known way to convert it into a 4x writer. Similar speculation has been made about the CDR-200 and CDR-400, and in fact some people have claimed success. Learn all about R621 at http://www.geocities.com/SiliconValley/Bay/7023/index.html (or http://home.t-online.de/home/christoph.dittenberger if you prefer German). It may also be possible to convert a 2260 into a 4260 with the same method, as well as the 2216 into a 4416.
Yamaha CDR-100 and CDR-102 units have problems doing digital audio extraction on some discs. See section (4-19).
Yamaha CDR-100s with firmware version 1.08 may experience problems when recording audio (e.g. a click at the end of tracks recorded with the "copy prohibit" flag set to "off"). Upgrading to version 1.10 is recommended. Since the CDR-100 and CDR-102 units don't have flash ROM (and apparently the upgrade involves more than just changing a ROM chip), the drive needs to be sent back to the dealer for the upgrade.
The CDR-100 reportedly works best when writing in 4x mode, and may produce poor results when used to write at 2x or 1x.
The current firmware versions for the older Yamaha drives is v1.12 for the CDR-100 and v1.01 for the CDR-102. The change was to "allow mastering in Blue Book specs". If you aren't having problems, don't get the upgrade. The Yamaha CDR-400 is somewhere around 1.0g.
The CDR-400 is flash upgradeable, and supports packet writing. The tray on the CDR-400 has been described as "flimsy". The tray eject moves quickly for the first half and then slows considerably; this is normal.
The CRW-4001/CRW-4260 runs rather hot. External units or extra cooling fans are recommended.
Some older Yamaha models apparently don't do disc-at-once recording. However, they will do session-at-once (SAO), which is as useful for most things and essential for multisession mixed audio and data discs. With the right software this isn't a problem.
CAVEAT EMPTOR - Yamaha CDR-200/CDR-400. Reports of units breaking down after a few months have been persistent. It appears that, unless the units are kept well-cooled, they will start rejecting discs after a month or two of use. The drives work very well otherwise, and one customer was told that the CDR-400AT model was a sturdier version.
(It may be possible to fix the drive by tightening some screws and adjusting some poorly-seated heat sinks on chips.)
CDW-900E (2x2/3MB:SCSI)The CDW-900E has a separate connector that allows multiple "slave" drives to be daisy-chained, allowing multiple CD-Rs to be written in parallel.
CRX100E/CH (24x4x2/1MB;IDE) and CRX100E/X (6x4x2/1MB;USB)
CRXP-90MU (24x24x10/8MB;USB2.0, also reads DVDx8, portable)
CRX1600L "i.LINK" (32x12x8/4MB;FireWire)
CRX175A/A1 (40x24x10/2MB;IDE, /A2 is USB)
DRX120A (32x12x10/2MB;IDE, DRX120L is ext. FireWire, also writes DVD+R/RW)
DRU128A (32x12x10/2MB;IDE, also writes DVD+R/RW)
DRU500A (32x24x10/8MB;IDE, also writes DVD+R/RW and DVD-R/RW)
DRU510A (32x16x8/8MB;IDE, also writes DVD+R/RW and DVD-R/RW; UL is USB2.0)
CRX195A1 (48x40x12/2MB;IDE, CRX1950U is external USB2.0)
CRX210A1 (48x48x12/2MB;IDE, CRX2100U is external USB2.0)
MPD-AP20U (24x24x10/?MB;USB2.0/1.1, also reads DVDx8 and plays DVD to TV)
CRX300A (48x48x24/2MB;IDE, reads DVDx16)
CRX320A/U (52x52x32/2MB;IDE, reads DVDx16)
[ There are also a number of DVD recorders that also record CD-R/RW ]
The Spressa 9211 is a 920 in an external case, the 9411 is a 940 in an external case, and the 9611 is a 926. The 940S drive is actually a 924S; the 940S designation refers to the complete bundle (software, cables, etc). Looks like each unit can be referenced by three different numbers.
Some people have criticized the CRX100E for being unable to write more than about 78 minutes on an oversized (e.g. 80-minute) blank, and being unable to "overburn" a disc without resorting to swap tricks. It appears that firmware v1.0n removes this limitation.
Some older Sony drives have a special "recover" feature, accessible from programs like Easy-CD Pro '95. This allows recovery of the CD-R media after certain classes of failed writes.
All Sony drives can do packet writing.
Firmware for some models can be hard to find.
CAVEAT EMPTOR - CDU926 and CDU928. Believe it or not, the CDU926 and CDU928 don't support disc-at-once recording (see section (2-9) for a description). Instead they use "variable-gap track-at-once", which allows TAO audio recordings with barely perceptible gaps between tracks. Some popular software packages aren't as useful when disc-at-once isn't available, so people considering these drives should carefully consider how they plan to use them.
(All other Sony units do support DAO.)
See http://tech.smartandfriendly.com/ (some ROM upgrades)
CDR1002 (2x2/1MB;SCSI, based on the Sony CDU920S)All models are recorders built by major manufacturers, repackaged and supported by Smart & Friendly.
CDR1004 (4x2/512K;SCSI, based on the Yamaha CDR-102)
CDR2004 (4x2/1MB;SCSI, based on the Sony 940S)
CDR2006 "Pro" (6x2/512K;SCSI, based on the Sony 926S)
CDR2006 "Plus" (6x2/1MB;SCSI, based on the JVC XR-W2020)
CDR4000 (4x4/512K;SCSI, based on the Yamaha CDR-100)
CDR4006 (6x4/2MB;SCSI, based on the Yamaha CDR-400)
CD-RW226 "Plus" (6x2x2/1MB;SCSI, based on the JVC XR-W2042)
CD-RW426 (6x4x2/2MB;IDE, based on the Yamaha CRW-4001/4260)
CD SpeedWriter 4012 (12x4/1MB;SCSI, based on the Teac CD-R55S)
CD SpeedRacer (16x4x4/2MB;SCSI, based on the Yamaha CRW-4416S?)
CD Racer 2x2x24 (24x2x2/2MB;IDE, based on the JVC XR-W2080)
CD SpeedWriter Plus (24x4x2/2MB;IDE (SCSI for ext), based on the JVC XR-W4080)
CD TurboWriter (24x6/2MB;SCSI, based on the Teac CD-R56S)
CD Rocket 8020 (20x8/2MB;SCSI, based on the Sanyo/Caravelle CRD-R800S)
CD Pocket RW (20x4x4/2MB;PCMCIA-2, based on ??)
CD Rocket RW (20x8x2/2MB;SCSI, based on ??)
CD SpeedWriter RW (24x4x2/?MB;SCSI, based on JVC XR-4424?)
CD TurboWriter RW (24x6x4/2MB;SCSI, based on Ricoh 7060A?)
CD CpeedWriter 32 (32x4x4/2MB;SCSI, based on Teac CD-W54E)
CD Rocket Mach 12 (32x12x4/4MB;SCSI, based on Sanyo CRD-RW2?)
CAVEAT EMPTOR - the company apparently went bankrupt in mid-May 2000. See the article at http://www.zdnet.com/zdnn/stories/news/0,4586,2597858,00.html. The web page was still running as of August 2000, but got changed to a pointer to justdeals.com after JustDeals bought up S&F's inventory.
See http://www.geocities.com/SiliconValley/Bay/6893/ (2600/3600)
CDD522 (2x2/2MB;SCSI)The CDD521 (2x2/256K) is an ancient model; if you use one, the firmware upgrade is strongly recommended (but nearly impossible to find these days). Some information that may be of use to CDD521 owners can be found at http://www.fadden.com/doc/cdd521faq.txt.
CDD3600 (6x2x2/1MB;SCSI) and CDD3610 (IDE)
CDRW200 (24x2x2/2MB;IDE, based on JVC XR-W2080? repackaged CDD3801?)
CDRW400 (16x4x4/2MB;IDE, based on Yamaha 4416E)
The Omniwriter/26 and /26A appear to be repackaged Ricoh 6200 and 6200I OEMs. In Europe, the 3600 is packaged in a kit as the PCA350RW, the 3610 as the PCA362RW, and the 3610 with a parallel-port interface comes as the PCA363RW. The CDRW400 might be packaged as the PCA460RW.
The CDD522 does not support reading of subcode-Q data. The CDD521, CDD522, and Kodak-labeled PCD225 have a sensor that can read the barcode data from the inner ring on a CD.
See the HP section for comments about the CDD2000 firmware. The firmware is kept in flash ROM, so it can be updated with software obtainable over the net. You should be at version 1.25 or later for best results.
Digital audio extraction may not work correctly at higher than 2x on the CDD2600, especially near the end of the disc. Philips has acknowledged that audio CDs and packet-written CDs may not read correctly at 6x, but many users have had problems at 4x as well. It may also suffer from the block offset problem described in section (4-19). The CDD2600 supports packet writing, but is NOT flash upgradeable.
The CDD2600 may share the HP 6020i's difficulties with pressed CD-ROMs that have a small amount of data on them.
The initial release (firmware v1.0) of the 3610 was unable to create audio discs reliably using disc-at-once recording. Firmware v2.02 fixed this and some other problems.
Philips' drives, notably the CDD2600, have been shown to hang on some Amigas if SCSI disconnect is enabled and you try to read the session information from a multisession CD. Philips does not believe this problem happens on PCs, and consequently has declined to investigate further. If you are experiencing hangs when examining multisession CDs, try turning SCSI disconnect off for the CD recorder.
CAVEAT EMPTOR - CDD2000. Some users of Philips CDD2000 and derivative units (like the HP4020i) have reported that the drives went bad over a short period of time, often 1 to 3 months. While these cases represent the minority of users, reports have been persistent. People with the technical skills (and bravery) required to replace a spring and/or lubricate inside the unit have reported good results (see section (4-10) for details). If you buy a CDD2000-based unit -- of which there are many -- be sure the dealer or manufacturer is aware of this problem and is willing to fix or exchange the drive should problems arise.
A class-action lawsuit was filed against Philips on behalf of owners of the CDD2000 and CDD2600. The case was eventually settled, with Philips agreeing to compensate the members of the class, either by replacing the defective drive with a newer model (which, unfortunately, was IDE instead of SCSI) or paying money to those who had bought a replacement drive and could document the expense.
4020i (4x2/1MB;SCSI, based on the Philips CDD2000) (#C4324)The 7110 is identical to the 7100, but comes with an extra piece of software and is only available in the USA. The 7200 is a 7100 with updated firmware (2.x) and Easy CD Creator included. The 8110 is the same as the 8100 but with a bunch of extra software; ditto for 8210 vs 8200. The 6020ep appears to be the external SCSI drive with a parallel-to-SCSI converter. It's usable as a SCSI device as well. The 71XXe drives are 71XXi drives with a parallel-to-IDE converter. The 7500 series is often packaged as 7550 or 7570 (though these are now listed as having only 1MB of buffer, so it's not clear what's going on).
6020 (6x2/1MB;SCSI, based on Philips CDD2600; i)nt, e)xt, p)arallel) (#C4325)
7100i/e (6x2x2/1MB;IDE, based on Philips CDD3610; 'i' is IDE (#C4353A),
'e' is parallel (#C4358A))
7500i/e (24x2x2/2MB;IDE, see note below, parallel 'e' model is 6x2x2)
8100i (24x4x2/1MB;IDE, based on the Sony CRX100E)
8200i/e (24x4x4/2MB;IDE, based on a Sony CRX120E; 'e' is 6x-read USB)
8250i (24x4x4/2MB;IDE, see note below)
9100i (32x8x4/4MB;IDE, see note below)
9200i (32x8x4/4MB;SCSI, based on Sony CRX140)
9300i (32x10x4/4MB;IDE, based on Sony CRX145E)
9500i (32x12x8/4MB;IDE, based on Sony CRX160E?)
9600i (32x12x8/4MB;SCSI, based on Sony CRX160S?)
9700i (40x16x10/8MB;IDE, based on ??)
9900i (32x12x10/2MB;IDE, reads DVDx8, based on ??)
cd12i (32x12x10/2MB;IDE, based on ??)
cd16i (40x16x10/2MB;IDE, based on ??)
cd24i (40x24x10/2MB;IDE, based on ??)
dvd100i (32x10x12/2MB;IDE, writes DVD+RW, based on Philips ??)
cd52i (52x52x32/?MB;IDE, 'e' is external)
There are indications that HP shipped two different drives as the 8250, both with the same read and write speeds. The first was the Philips CDD4201, identifiable by about 18 tiny horizontal indentations along the bottom of the face, with a hinged "drawbridge" loading door. The second was the Sony CRX120E, which has 4 horizontal indentations along the bottom of the face, and no hinged door. The popular consensus is that the Philips versions are problematic. It may be possible to tell the boxes apart using a code on the barcode label: C4464A for Philips, C4464B for Sony. (It appears there may even be a third variety: HP is rebadging Mitsumi 4804TE in their Pavilion 6648C computers. There doesn't appear to be an HP model number associated with the drive though, so it may not actually be sold as an 8250.) It now appears that some 8250i drives are 32x4x4/4MB; these are actually Sony CRX140E drives (32x8x4/4MB) with firmware that limits them to 4x recording. It has been reported that, if you can get the HP9100i firmware onto the drive, it will record at 8x.
There are similar indications for the 7500, which appears to have originally been a JVC XR-W2080, but is now a Sony CRX100E with a reduced maximum write speed. Drives based on the Sony mechanism can reportedly be flashed with the Sony CRX100E firmware update and upgraded to 4x recording.
Most 9100 uints are based on the Sony CRX140, but there are indications that units identifying themselves as "9100b" is actually a Goldstar CED-8080B.
The initial release of the 7100/7110 was unable to create audio discs reliably with disc-at-once recording. The 2.02 firmware upgrade fixes the problem.
Some people have criticized the 8100i (same as Sony CRX100E) for being unable to write more than about 78 minutes on an oversized (e.g. 80-minute) blank. It appears that the Sony v1.0n firmware upgrade removes this limitation, but the upgrade was never made available for the HP drive. Some users have had success flashing the drive with a "hacked" version of the Sony firmware, but this can be dangerous (see warnings in section (5-24)). You can also just use the Sony firmware, but that causes the front LEDs to stop working.
If you are having trouble getting the 7100e to work with your parallel port, see http://www.hp.com/isgsupport/cdr/tech/7100/par95.html for some important configuration advice. If your BIOS is configured to use address 03BCh, you should change it to 0378h or 0278h.
It appears that discs written with a 7110 can't be read on a Toshiba XM6002B. Other models of CD-ROM drives, including other Toshiba models, work fine. CD-Rs written on other CD recorders work fine with the Toshiba. The 3.01 firmware upgrade fixes this.
The HP 4020i got off to a rough start because of buggy firmware and problems with the AdvanSys SCSI controller shipped with the drive. Four firmware upgrades have been made available so far (v1.20, v1.25, v1.26, and v1.27), and most but not all problems with the firmware have been eliminated. HP recommends that users with the v1.20 or later firmware who aren't having problems should NOT get the upgrade. Contact HP tech support for more information.
The comments about digital audio extraction problems and the CDD2600 apply to the 6020i as well. Unlike the CDD2600, the 6020 apparently does not support packet writing. The firmware is not flash upgradeable. (As it happens, the SCSI ID string *can* be changed, and it *is* possible to make the unit think it's a CDD2600. A representative from Adaptec has warned that the procedure could cause problems later on, however.)
The 6020 with v1.07 firmware also has trouble reading some pressed CD-ROM discs, notably single-track CD-ROMs with less than 27MB of data.
An unofficial HP 4020i FAQ maintained by Greg Volk can be found at http://www.cd-info.com/CDIC/Technology/CD-R/HP-FAQ.html.
Drivers, software, and firmware upgrades are available from ftp://ftp.hp.com/pub/information_storage/surestore/cd-writer/.
The 7100/7110 firmware upgrade is available here: http://www.hp.com/isgsupport/cdr/70index.html
IMPORTANT - 7100/7110. The 7100/7110 drew a lot of fire because it shipped with DirectCD (packet-writing software), a CD Copier, and an audio CD creator. It didn't include premastering software for data CDs. Because packet-written CDs can't be read on all operating systems or all CD-ROM drives, the inability to create plain Level 1 ISO-9660 discs was a problem for some users. People who buy this drive should expect to buy additional software. The software bundled with the 7200 was more wisely chosen.
CAVEAT EMPTOR - 4020i. See the notes on the CDD2000 in the previous section. Also, the AdvanSys controller continues to cause problems for some users, which is made worse by HP's refusal to support people who try to use a different card. The best approach seems to be to try the card and stick with it if it works, otherwise buy an Adaptec board (e.g. the 1522A) and use it with that. There may be a newer rev of the AdvanSys board.
A few 4020 users have reported that, after getting lots of "-24 - Target aborted" errors with jarnold's software, they successfully resolved their problems by getting a new drive from HP.
A class-action lawsuit was filed against HP (for the HP4020i and HP6020i) by the same people who filed the suit against Philips -- the drives were repackaged versions of the CDD2000 and CDD2600.
RF4100 (2x2/1MB+;SCSI, based on Philips CDD522 but with different firmware)The RF4102 is an RF4100 with more memory.
CDR4220 (4x2/1MB;SCSI, based on the Philips CDD2000)
CDR4240 (4x2/1MB;SCSI, based on the Panasonic CW-7501)
CDR-4400 (4x2/512K;SCSI, based on the Yamaha CDR-100)
CDR480 (8x4/1MB;SCSI, based on the Panasonic CW-7502)
The RF4100 does not support disc-at-once recording.
See http://www.kodak.com/ [ no CD recorder info? ]
PCD200 (?x2/256K;SCSI)The Philips CDD522, Kodak PCD225, and Kodak PCD600 will interface with the Kodak Disc Transporter, which supports unattended duplication of up to 75 CD-Rs, making it a useful combo for CD-R production.
PCD225 (2x2/2MB;SCSI, based on the Philips CDD522)
PCD240 (4x2/1MB;SCSI, based on the Philips CDD2000)
4801 (??;IDE, based on the Mitsumi 4801??)
XR-W1001 (1x1/64K;SCSI)[ JVC stopped selling CD recorders somewhere around 2002. ]
[on the JVC web site, no model number?] (12x4/1MB;IDE)
The drives are sometimes sold with model numbers that have 2 added, so XR-W2010 might appear as XR-W2012, XR-W2020 as XR-W2022, and XR-W2080 as XR-W2082. The XR-W2626 appears to be an XR-W2020.
The drives often come bundled with JVC "Personal Archiver" or "RomMaker" software. The XR-W2010 and XR-W2020 also come with "FloppyCD" packet-writing software.
JVC only provides support for drives purchased directly from them, but firmware updates can be found at http://www.jvcinfo.com/service/firmware.htm. If you don't buy a JVC drive from JVC, make sure your vendor provides a warranty.
If you are getting "servo tracking error", "seek error", or "track following error" messages with an XR-W2010 or XR-W2020, your drive may need to be opened up and lubricated. Step-by-step instructions for doing so can be found on http://www.smial.prima.de/old/howtoget.htm. If you're not quite up to that, try turning the drive off and leaving it off until right before you're ready to burn. Some units have trouble when they get warm.
Several users have reported difficulty installing the XR-W2020, but the troubles appear to stem from the SCSI card bundled with the drive rather than the drive itself.
CAVEAT EMPTOR - XR-W2010. Firmware version 1.51 has some serious flaws that can cause problems when using the drive as either a writer or a reader. The v2.05 update fixed most of the problems, but some conflicts with 3rd-party software remained, so the update was withdrawn. Until these problems are fixed, this drive should only be used with the JVC software, and should not be used as a reader. Power-cycling the unit (i.e. powering it off and back on) immediately before a write may cure some problems. For examples and some tests, see http://www.fadden.com/doc/jvc-prob.txt.
While there are a large number of people who are using these drives without problems, one person affiliated with a CD-R software company referred to the XR-W2010 as their "#1 tech support nightmare".
CAVEAT EMPTOR - XR-W2020. The mechanism appears to have the same problems with lubrication as the XR-W2010. After several months of successful use, the unit will start returning "tracking error" messages.
RCD-202 (?x1/64K;SCSI, based on the JVC XR-W1001)The -1000, -5020, and -5040 models are flash ROM upgradeable.
RCD-1000 (2x2/1MB;SCSI, based on the JVC XR-W2001)
RCD-5040 (4x2/1MB;SCSI, based on the JVC XR-W2010)
RCD-4X4 (4x4/1MB;SCSI, based on the Teac CD-R50S)
RCD-1000 units shipped after Sept 1995 can do audio extraction if they have firmware v2.35 or later. An upgrade is available from their BBS.
If you are getting "servo tracking error", "seek error", or "track following error" with a 5040, see the notes in the JVC XR-W2010 section.
CAVEAT EMPTOR - all drives. Pinnacle customer support is reported to be almost nonexistent, except for some recent tech support via e-mail. Many owners of the RCD-5040 are perfectly happy with their drives (see the caveat on the JVC XR-W2010), but most of the stories about Pinnacle's product support are negative.
Pinnacle earned a bad reputation after shipping drives with buggy firmware, a poorly ventilated enclosure, and bad customer support. Some owners of the RCD-1000 have gotten the unit to work, others have given up in despair.
RS-9200CD (?x1/1.2MB;SCSI)[ Ricoh stopped selling CD recorders somewhere around 2002. ] The MP-6200 uses a tray, the MP-6201 uses caddies and has a 2MB buffer.
MP-6200 (6x2x2/1MB; 'S' is SCSI, 'A' or 'I' is IDE)
MP-7040 (20x4x4/2MB; 'S' is SCSI, 'A' is IDE)
MP-7060 (24x6x4/2MB; 'S' is SCSI, 'A' is IDE)
MP-8040SE (20x4x4/2MB;PCMCIA-2(SCSI), battery-powered)
MP-9060A (24x6x4/2MB;IDE, reads DVDx4)
MP-9120A (32x12x10/2MB;IDE, reads DVDx8)
MP-9200A (40x20x10/2MB;IDE, reads DVDx12)
The RS1060C does not support disc-at-once recording, reading of digital audio, or subcode-Q data. (One user reported that his RO1060C *could* read digital audio, but the drive took a little convincing. Another user says that it has always been supported, but not documented, so it can be done with the right software, e.g. CDDA v1.5.) The RS-1060C is the RO-1060C in an external case.
The RS-1420C is flash upgradeable (though it can be a little tricky since there are different variants of the drive, and each requires a different ROM image). It does not support packet writing. Most of the commercial versions come with a 2MB buffer (the last digit of the firmware version will be 0, 1, or 2, indicating 512K, 1MB, and 2MB, respectively).
The firmware on the flash-upgradeable MP-6200 should either be at v2.20 or later. Version 1.0 had several problems, version 2.0 didn't get along so well with DirectCD 2.0, and version 2.03 had some DAE issues.
Firmware upgrades are available from Tom Varghese's page listed above (arrakis-ttm.com) and http://www.ricoh.co.jp/cd-r/cgi/e-/version.html.
The MP-6200 "red/green" problem, where the drive starts having trouble accepting media, and sits there flashing red and green, appears to be caused by a buildup of oil on the drive's spindle clamp. See the arrakis-ttm.com site for details.
Some people have found that the MP7040/7060 will start to "stick" after a while, resulting in consistent write errors at roughly the same spot every time. Some people have found that lubricating the drive helps. This is a dangerous procedure, and should not be attempted unless all other possibilities have been exhausted. Details can be found on http://www.don.cohoon.net/ricoh/ricoh.html.
See http://www.pioneerusa.com/cds.html [ mass replication ]
See http://www.pioneer.co.jp/ [ if you can read Japanese ]
DW-S114X (4x4/1MB;SCSI)The PDR-05 is an audio CD-R recorder, described in section (5-12).
Does not support disc-at-once recording. Mainly sold in large jukebox systems.
CDS615E (2x2/1MB;SCSI, based on the Sony CDU920S)The CD-R2 is the CDS615E in an external case. The CD-R2x4 might be the external version of the CDS620E. The CD-R2x6 probably has a name like CDS640E, but it's not listed as such on their web site.
CDS620E (4x2/1MB;SCSI, based on the Sony CDU924S)
CD-R2x6 (6x2/512K;SCSI, based on the Sony CDU926S??)
DisKovery 650 CD-R (2x2/1MB;SCSI, based on the Sony CDU920S)As of the middle of 2003, Optima was busily suing CD-R software manufacturers and resellers over (among other things) US patent #5,666,531. This patent, filed in April of 1995, appears to cover packet writing.
DisKovery 1300 CD-R (6x2/512K;SCSI, based on the Sony CDU960S?)
Optima CDWriter (6x4x2/2MB;SCSI, based on ??)
CR-2200CS (2x2/4MB;SCSI, based partly on the Philips CDD2000)In all unit designations, 'C' means caddy, and 'T' means tray, 'S' is SCSI, and 'E' is IDE.
CR-2201CS (same as CR-2200CS but with 2x2/1MB)
CR-2401TS (4x2/1MB;SCSI, based on the Philips CDD2000)
CR-4802TE (8x4x2/2MB;IDE) and CR-4802TU (USB)
CR-480ATE (40x32x12/2MB;IDE, sometimes referred to as 48xA)
The devices based on the CDD2000 are flash upgradeable (you should be able to use Philips CDD2000 images).
CAVEAT EMPTOR - CR-2600TE and CR-2801TE. These drives do not support disc-at-once recording. Like the Sony 926 and 928 units, they claim to support track-at-once with nearly imperceptible gaps instead. Ahead's Nero can reportedly do this with the CR-2801TE.
The CR-4801TE with firmware 2.01 and later supports DAO recording. Earlier versions do not. If your recording software doesn't believe that the drive is capable of DAO, you may need to update the software to a version that is aware of the changes in the firmware update.
Later drives, such as the 4802TE, do support DAO.
CDRW8424 (24x8x4;SCSI)[ DynaTek reportedly went out of business. However, the UK site seems to be alive and well. ]
CDM240J (4x2/512K;SCSI, based on the JVC XR-W2010)
CDM400 (4x4/512K;SCSI, based on the Yamaha CDR-100)
CDE260R (6x2x2/1MB;SCSI, based on the Ricoh 6200S)
Older CDM240 units were based on the Yamaha CDR-102. Since the Yamaha CDR-100 is no longer being made, chances are the CDM400 is now a different unit as well.
They also sell the CDM4000, which is a stand-alone CD burner.
PlayWrite 2000 (2x2/1MB;SCSI, based on the Sony CDU920S)
PlayWrite 2040 (4x2/512K+;SCSI)
PlayWrite 4000 (4x4/512K;SCSI, based on the Yamaha CDR-100)
PlayWrite 4001RW (6x4x2/2MB;IDE, based on the Yamaha CDR4001t)
PlayWrite 2060R (6x2x2/1MB;SCSI, based on the Ricoh 6200S)
Model is the Express Writer. There are no apparent model numbers. They used to sell the "old one" (2x2/1MB, based on a Pinnacle (i.e. JVC) drive), more recently they sold the "new one" (4x2/?).
See http://www.micronet.com/HTDOCS/products.html#cdr [ site gone? ]
MasterCD Plus 4x4 (4x4/512K;SCSI, based on the Yamaha CDR-100)
MasterCD Plus 4x6 (6x4/2MB;SCSI, based on the Yamaha CDR-400)
MasterCD Plus 4x12 (12x4/1MB;SCSI, based on the Teac CD-R55S)
PCDR-4X (4x4/512K;SCSI, based on the Yamaha CDR-100)
See http://www.grundig.com/ [mostly in German]
CDR100IPW (4x2/1MB;SCSI, based on the Philips CDD2000)
PX-R24CS (4x2/512K;SCSI, a cousin of the Ricoh 1420C)For all units, 'C' indicates caddy, 'T' indicates tray, 'S' is SCSI, 'A' is ATAPI, 'U' is USB.
PX-W8432T (32x8x4/2MB;IDE), also SCSI PX-W8432Ti/SW with 4MB
PX-W1210TA (32x12x10/2MB;IDE), also SCSI PX-W1210TS with 4MB
PX-W2410TA (40x24x10/4MB;IDE, also 'U' portable USB)
PX-208U (24x8x8/2MB;USB2.0, reads DVDx8, portable)
PX-320A (40x20x10/2MB;IDE, reads DVDx12)
PX-W4824TA (48x48x24/4MB;IDE, TU model is external USB2.0)
All units are flash upgradeable. All units except the PX-R24CS support packet writing.
Users having trouble with the PX-R412C should try turning synchronous transfer off for that drive.
There appears to be an issue with the Plextor PX-320A and a SiS IDE chipset. Using the DMA jumper to change the Plextor drive from UltraDMA to multi-word DMA fixes the problem.
CW-7501 (4x2/1MB;SCSI)Panasonic is part of Matsushita, so the units may also be sold under the Matsushita label.
All units are flash-upgradeable. The CW-7501 should be at 2.0 or greater, and the CW-7502 should be at vX.10 or later (1.10, 3.10, or 4.10 depending on which recorder variant you have; check your current version). Upgrades are available from http://www.acscompro.com/ (click on "Support") [site was down as of May 2002?].
NOTE: there is a known conflict with the Diamond FirePort 40 and the Panasonic CW-7502 CD-R drive. You should upgrade the 7502 firmware to the latest (www.acscompro.com/support/s_cdr.htm), upgrade your FirePort 40 drivers (http://www.diamondmm.com/products/drivers/fireport.html), and add "DisableAutoReqSense=1;do_SCAM=0;" to the FirePort driver (go into the Win95 device settings, select the host adapter, click on Properties, and select the Settings tab).
This problem may affect other NCR/Symbios Logic-based SCSI cards as well. Falling back to the original (1.01) NCR SCSI drivers that come with Win95 should fix the problem.
NOTE: the 7502/7503 units may have a problem with writing near the end of 80-minute discs. The problem is fixed by a firmware upgrade. If you get errors reading data stored near the end of the disc (e.g. errors creating a disc image from a full 80-minute CD or CD-ROM), make sure you have the latest firmware.
CD-R50S (4x4/1MB;SCSI)Apparently the CD-R50S needs to be at firmware 1.0E or later to do quad-speed writing reliably. Power calibration is done via a lookup table rather than adjusted dynamically, so a flash upgrade may be required before some brands of media will work.
CD-W512 (32x12x10/4MB; 'E' is IDE, 'S' is SCSI)
CD-W516 (40x16x10/2MB; 'E' is IDE)
CD-W540E (48x40x12/8MB;IDE, F540 is external USB (6x4x4) or USB2.0)
DW552G (52x52x32/?MB;IDE, reads DVDx16)
The CD-R50S and CD-R55S appear to use the same command set as the JVC XR-W2010.
http://www.teac.co.jp/dspd/download/firmware/cd-r55s/updater.html has a nice HTML page about the CD-R55S upgrade.
See http://www.wpinet.com.sg/ [site gone?]
CDR-432 (4x2/1MB;SCSI, based on the Philips CDD2000)The CD-R 622 does not support disc-at-once recording. According to the CDRDAO "readme" file, it is possible to upgrade the 622 (and its Memorex cousin) by writing the D4.0 ROM image for the CRW-1622 to a 27c020 PLCC EPROM and replacing the socketed ROM chip in the drive.
CD-R 622 (6x2/1MB;IDE)
CD-R 632P (6x2/1MB;SCSI, based on the Philips CDD2600)
The CDRW-622 supports packet writing, and is flash upgradeable.
2040R (4x2/512K;SCSI, based on the Ricoh RS-1420C)Many users have had trouble installing the AdvanSys SCSI card that is bundled with this unit. Most of the problems can be corrected by enabling PnP installation, which is disabled by default.
CDR2000 (2x2/512K;SCSI, based on the Ricoh RS1060C)Creative sold several drives with the 32x8x4 rating, starting with the Plextor-based 8432. According to some information [formerly at] http://www.ping.be/satcp/writer04.htm, the 8433, 8435, 8438, and 8439 are similar but different devices. Looks like they did something similar with the 32x12x10 drive.
CDR2224 (24x2x2/2MB;IDE, based on JVC XR-W2080?)
CDR4210 (4x2/1MB;SCSI, based on the Panasonic CW-7501)
CDR4224 (24x4x2/2MB;IDE, based on a JVC XR-W4080)
CDR?? "CD Studio" (24x4x4/2MB;IDE, based on ??)
CDR6424 (24x6x4/2MB;IDE, based on Ricoh 7060A)
CDR8432 (32x8x4/2MB;IDE, based on PX-W8432T; also 8433/8435/8438/8439)
CDR8433 (same as 8432, based on Panasonic CW-7585)
CDR8435 (same as 8432, based on Samsung SW-208)
CDR8438 (same as 8432, based on Samsung ??)
CDR8439 (same as 8432, based on Panasonic CW-7586)
CDR121032 #1 (32x12x10/2MB;SCSI, based on Plextor PX-W1210)
CDR121032 #2 (32x12x10/2MB;SCSI, based on Lite-On LTR-1210)
CDR161040 (40x16x10/2MB;IDE, based on ??)
CDR241040 (40x24x10/2MB;USB or FireWire, based on ??)
CD-RW 52.24.52 (52x52x24/?MB;IDE)
CD-RW 52-32-52x (52x52x32/?MB;IDE)
Generally speaking, reading the retail box won't tell you what's inside.
CR-622 (6x2/1MB;IDE, based on the Wearnes CD-R 622)
CRW-1622 (6x2x2/1MB;IDE, based on the Wearnes CDRW-622)
CRW-2642 (6x4x2/2MB;IDE, based on the Yamaha CRW-4260??)
CDRW-2216 (16x2x2/1MB;IDE, based on the Yamaha CRW-2216E)
CDRW-2224 (24x2x2/2MB;IDE, based on JVC XR-W2080?)
CDRW-4206-USB (6x4x2/2MB;USB, based on ??)
CRW-4224 (24x4x2/2MB;IDE, based on JVC XR-W4080?)
CDRW-8220 (20x8x2/2MB;SCSI, based on ??)
CDRW-12432 (32x12x4/2MB;IDE, based on ??)
"32X CD ReWritable Drive" (40x32x12/?MB;IDE)
"40X CD ReWritable Drive" (48x40x12/2MB;IDE)
"48X CD ReWritable Drive" (48x48x12/2MB;IDE)
"48Xv2 CD ReWritable Drive" (48x48x24/2MB;IDE)
"52X CD-ReWritable Drive" (52x52x24/?MB;IDE)
Hi-Val doesn't build CD recorders. They repackage and provide support for recorders built by others. The actual model you get will vary (Wearnes, Ricoh, Philips, JVC, Mitsumi, and others have been reported).
CR-622 (6x2/1MB;IDE, based on the Wearnes CD-R 622)The CRW-1622 often came bundled with NTI's software, but the version included didn't work correctly. Upgrading to a more recent version of the software (http://www.ntius.com/) resolved the problems.
CRW-1420C (6x2/512K;SCSI, based on the Ricoh 1420C??)
CRW-1622 (6x2x2/1MB;IDE, based on the Wearnes CDRW-622)
CDRW-2216 (16x2x2/1MB;IDE, based on the Yamaha CRW-2216E)
See http://www.traxdata.com/ [ site requires Flash ]
CDR4120 (12x4/1MB;SCSI, based on the Teac CD-R55S)The CDRW2260 "Pro" may also use a Philips CDD3600?
CDRW2260 "Pro" (6x2x2/1MB;SCSI, based on the Yamaha CRW-2260)
CDRW2260 "Plus" (6x2x2/1MB;IDE, based on the Philips CDD3610?)
CDRW-4260 "Pro" (6x4x2/2MB;SCSI, based on the Yamaha CRW-4260)
CDRW-2224 "Plus" (24x2x2/?MB;???, based on Philips CDD3801?)
CDRW-4424 "Plus" (24x4x4/2MB;IDE, based on the Philips CDD4201?)
CDW6206A (6x2x2/512K;IDE)A user who was getting nothing but power calibration complaints with the CRW1032A and firmware 7.EZ found a laser power adjustment tool in the 7.GZ update from the www.acercm.com site. The North American version reportedly doesn't come with the tool, but it may not be needed.
CRW2410MR (32x24x10/?MB; external USB2.0)
CRW3210A (40x32x10/?MB;IDE 'AI' is USB2.0)
CRW4012P (48x40x12/?MB;IDE, 'EU' is USB2.0)
CRW5224P (52x52x24/2MB;IDE, 'WU' is USB2.0)
WT4046 (6x4x2/2MB; "EI" model is IDE)
WT2036 (6x2x2/1MB; "EI" model is IDE)
WT2082 (20x2x2/4MB;SCSI, "EXT" is external, based on ??)
WT2444EI (24x4x4/2MB;IDE, based on the Philips CDD4201?)
WT3244EI (32x4x4/2MB;IDE, based on ??)
WT3284EI (32x8x4/4MB;IDE, based on Plextor PX-W3284?)
"Raptor" (32x12x10/4MB; "Red" is IDE; based on Sanyo CRD-BP1300P??)
"X-File" (32x12x10/2MB;IDE, reads DVDx8, writes DVD+RWx2.5)
"T-Rex" (40x16x10/2MB;IDE, based on Sanyo CRD-BP1400P??)
"SfinX 16" (40x16x10/8MB;IDE, also reads DVDx10)
"Storm 24" (40x24x10/?MB;IDE)
"Storm 32" (40x32x10/4MB;IDE)
"Frisby II" (40x40x12/2MB;USB2.0, portable)
"Storm 40" (48x40x12/4MB;IDE)
"Storm 48" (48x48x16/2MB;IDE)
"Storm 52" (52x52x24/2MB;IDE)
"Storm 52/3" (52x52x32/2MB;IDE)
[ See also the "Action" line of DVD/CD recorders. ]
BCE62IE (6x2x2/1MB;IDE, based on the Philips CDD3610??)The BCE62IPE is the BCE62IE with a parallel-port IDE converter.
CRD-R800S (20x8/2MB;SCSI)[ Sanyo stopped selling CD recorders somewhere around 2002. ]
CRD-BP4 (40x16x10/2MB;SCSI, also in 4MB)
CRD-BP1400P (40x16x10/2MB;IDE, also in 4MB)
CRD-BP1500P (40x24x10/2MB;IDE; 'U' is USB)
CRD-SBP15A (32x24x10/2MB;IDE, portable, for OEM only)
CRD-BP1500U40X (40x40x12/4MB;IDE, external is USB2.0)
Firmware v1.10 or later is highly recommended for the CRD-R800S. For some reason, the firmware update was only available on the "BURN-Proof" web site at http://www.sannet.ne.jp/BURN-Proof/. [ It doesn't seem to be there anymore. ]
It looks like Mirai Technologies (http://www.mirai-technologies.com/) resells these drives.
190100 (6x2x2/1MB;Parallel, based on the Ricoh MP-6200)All products are standard recorders combined with Micro Solution's parallel-port interface.
190120/190126 (6x4x2/?MB;Parallel, based on the Yamaha CRW-4261)
190127 (8x4x2/2MB;Parallel, based on the Mitsumi CD-4802TE)
224ei (24x2x2/2MB;IDE, based on the JVC XR-W2080)
226ei (6x2x2/1MB;IDE, based on the Philips CDD3610??)
428ei (8x4x2/2MB;IDE, based on the Mitsumi CR-4802TE)
428USB (8x4x2/2MB;USB, based on the Mitsumi CR-4802TU)
416si (16x4x4/2MB;SCSI, based on Yamaha CRW-4416S?)
448USB (8x4x4/2MB;USB, based on ??)
8824si (24x8x8/4MB;SCSI, based on Yamaha CRW-8824??)
8832ei (32x8x8/2MB;IDE, based on ??)
121032ei (32x12x10/2MB;IDE, based on Lite-On 32x12x10)
161040ei (40x16x10/8MB;IDE, based on CRW-2100?)
241040ei (40x24x10/2MB;IDE, also as USB)
Xtreme32 #1 (40x32x10/2MB;USB2.0 and USB (8x4x4))
Xtreme32 #2 (48x32x12/2MB;USB2.0 and USB (8x4x4))
Mach40 #1 (48x40x12/2MB;IDE)
Mach40 #2 (48x40x16/2MB;IDE)
Xtreme48 (48x48x12/2MB;USB2.0 and USB (8x4x4))
Mach52 (52x52x24/2MB;IDE, also available in USB2.0)
Blue Lightning52 (52x52x24/?MB;IDE)
[ Some DVD/CD recorders are also available. ]
ZipCD (24x4x4/2MB;IDE, based on the Philips CDD4201)
ZipCD external (6x4x4/2MB;USB, based on ??)
CD-RW Predator 8x4x32 FireWire (32x8x4/2MB;FireWire)
ZipCD 12/10/32 (32x12x10/2MB;IDE, based on Plextor PX-W1210T)
ZipCD 16/10/40 (40x16x10/2MB;IDE, based on ??)
CD-RW 24x10x40 USB (40x24x10/2MB;USB)
CD-RW 40x12x48 USB (48x40x12/2MB;USB2.0), also available as FireWire
CD-RW 48x24x48 USB (48x48x24/2MB;USB2.0)
CD-RW 52x24x52 USB (52x52x24/2MB;USB2.0)
CD-RW 52x32x52 USB (52x52x32/?MB;USB2.0)
CED-8041B (24x4x2/2MB;IDE)The first two *might* be based on the Sony 100/120 models. There are indications that, at the very least, the firmware is different (the Goldstar units reportedly can "overburn" discs, while the mentioned Sony units couldn't when these were released.)
GCC-4120B (32x12x8/2MB;IDE, reads DVDx8)
GCC-4320B (40x32x10/2MB;IDE, reads DVDx16)
GCE-4480B (48x48x24/2MB;IDE, reads DVDx16)
GCE-4520B (52x52x24/2MB;IDE, reads DVDx16)
CR1420C (4x2/512K;SCSI, based on the Ricoh RS-1420C?)
CRW620 (6x2/1MB;SCSI, based on ??)
CRW622 (6x2/1MB;IDE, based on Wearnes CD-R 622??)
CRS446U (6x4x4/1MB;USB, "crab shell")
CRW9420 (20x4x4/2MB;IDE, based on Ricoh MP-7040A?)
CRW9624 (24x6x4/2MB;IDE, based on Ricoh MP-7060A?)
DRW4624 (24x6x4/2MB;IDE, reads DVDx4)
RW5120A (32x12x10/2MB;IDE, reads DVDx2.4)
DVRW2412PRO (32x12x10/2MB, also reads DVD+RW)
CRW3248 (48x32x12/2MB;IDE, has an option for 8MB buffer)
[ Some DVD/CD recorders are also available. ]
SD-R1002 (24x4x4/2MB;IDE, reads DVDx4)
8/4/32 veloCD (32x8x4/4MB;IDE, based on ??)
12/10/32 veloCD (32x12x10/2MB;IDE, based on Plextor PX-W1210TA)
16/10/40 veloCD (40x16x10/2MB;IDE, based on Sanyo CRD-BP1400P?)
24/10/40 veloCD (40x24x10/2MB;IDE, also external USB2.0 and FireWire)
32/10/40 veloCD (40x32x10/2MB;IDE, based on Sanyo CRD-BP1600PN?)
40/12/48 veloCD (48x40x12/2MB;IDE)
48/16/48 veloCD (48x48x16/2MB;IDE, also external USB2.0)
52/24/48 veloCD (52x48x24/2MB;IDE)
LTR-0841 (32x8x4/2MB;IDE)Some of the drives appear to be based on Plextor units. It has been claimed that the LTR-0841 can be upgraded to an LTR-12101B with a firmware upgrade; see http://www.cdrinfo.com/Sections/Tips/oc_Lite-On.asp.
LTR-48125S (48x48x12/2MB;IDE, same as 48125W/48126S?)
[ Combo DVD/CD recorders are also available. ]
Customer support issues are deferred to the dealer.
There is an internal configuration program called "WSES" that can be used for testing drives and discs. Copies can be found on the web.
CDI CD00000 (20x4x4/2MB;SCSI)[ CenDyne was acquired by Genica Corporation in December 2003. The range of products and services appears to be much smaller than before. ]
CDI CD00001 (20x4x4/2MB;SCSI)
CDI CD00015 (20x4x4/2MB;SCSI)
CDI CD00016 (24x4x2/2MB;IDE)
CDI CD00017 (24x4x4/2MB;IDE)
CDI CD00018 (32x4x4/2MB;IDE)
CDI CD00023 (32x8x4/2MB;IDE)
CDI CD00028 (32x12x4/4MB;SCSI)
CDI CD00029 (24x6x4/2MB;SCSI)
CDI CD00030 (24x6x4/2MB;SCSI)
CDI CD00032 (24x6x4/2MB;SCSI)
CDI CD00036 (20x8/2MB;SCSI)
CDI CD00037 (20x8/2MB;SCSI)
CDI CD00038 (20x8/2MB;SCSI)
CDI CD00039 (20x8/2MB;SCSI)
CDI CD00040 (24x6x4/2MB;SCSI)
CDI CD00045 (32x12x4/4MB;SCSI)
CDI CD00047 (32x6x4/2MB;IDE)
CDI CD00055 (32x12x10/4MB;IDE)
CDI CD00056 (24x4x4/2MB;IDE)
CDI CD00057 (32x8x8/2MB;IDE)
CDI CD00059 (32x12x10/2MB;IDE, reads DVDx8)
CDI CD00063 (32x12x10/?MB;Firewire)
CDI CD00068 (20x4x4/?MB;PCMCIA)
CDI CD00086 (20x4x4/?MB;USB)
CDI CD00087 (40x16x10/?MB;IDE)
CDI CD00090 (40x20x10/2MB;IDE)
CDI CD00091 (40x24x10/2MB;IDE)
CDI CD00092 (20x4x4/2MB;PCMCIA or USB)
CDI CD00094 (40x24x10/2MB;FireWire)
CDI CD00102 (32x12x10/2MB;USB2.0)
CDI CD00103 (40x16x10/2MB;USB2.0)
CDI CD00104 (40x24x10/2MB;USB2.0)
CDI CD00107 (40x32x12/2MB;IDE)
CDI CD00116 (24x8x8/2MB;USB2.0)
CDI CD00117 (48x40x12/2MB;IDE)
CDI CD00118 (48x48x12/?MB;IDE)
CDI CD00122 (48x40x12/2MB;IDE)
CDI CD00123 (40x16x10/2MB;IDE, read DVDx12)
CDI CD00134 (40x40x12/2MB;USB2.0, top-loading)
CDI CD00137 (48x48x12/?MB;IDE)
CDI CD00154 (40x32x12/2MB;USB2.0)
CDI CD00167 (40x32x10/2MB;IDE, read DVDx12)
CDI CD00172 (48x40x12/2MB;USB2.0)
All models are recorders built by major manufacturers, repackaged and supported by CenDyne. In many cases the model numbers refer to slight changes in packaging (e.g. Windows vs Mac) or internal vs external variations of the same drive.
CenDyne has the distinction of using the least imaginative naming scheme of any distributor (the polar opposite of Waitec).
VST Portable CD-R/RW (20x4x4/2MB;FireWire;portable)
CRW-4012A (48x40x12/2MB;IDE, "-U" model is external USB2.0)
SN-308B (24x8x8/2MB;IDE, read DVDx8)
SM-308B (32x8x4/2MB;IDE, read DVDx8)
SM-316B (40x16x10/8MB;IDE, read DVDx12)
SM-332B (40x32x10/8MB;IDE, read DVDx12)
SM-348B (48x48x24/8MB;IDE, read DVDx16)
SW-252 (52x52x24/2MB;IDE, retail version has 8MB buffer)
[ Some DVD/CD recorders are also available. ]
APS "52x24x52 FireWire & USB CD-RW" (52x52x24/?MB;FireWire & USB2.0)[ APS was purchased by LaCie in 1998. ]
Products are repackaged drives from other manufacturers.
The MTBF (Mean Time Between Failures) on these drives is typically 50,000 to 100,000 hours, and they come with a 1 year warranty. Compare that to hard drives rated at between 500,000 and 1,000,000 hours with a 3 or 5 year warranty and that should give you some idea.
Most of the drives available today weren't meant for mass production of CD-Rs. The only exceptions are the venerable Philips CDD 522, Kodak PCD 600, and Sony CDW-900E.
Incidentally, MTBF is not an estimate of how long the drive will last. Rather, it's an estimate of the failure rate of the drives during the expected lifetime of the device. Once you exceed the expected lifetime, which is often on the order of a couple of years, the anticipated failure rate increases. If you have new drives with an MTBF of 25,000 hours, and you run 1000 units for 100 hours, you can expect to see four of them fail. It does NOT mean you can expect them to run for 2.8 years and then all fail at once.
If you're about to buy a computer system and are seriously thinking about buying a CD-R, here are some things to keep in mind. (See the next section if you're interested in Mac hardware instead of an IBM PC.)
CPU: buy a mid-range Pentium-class machine or better. In general it's a good idea to buy a fast machine, since systems tend to be outdated after a year and obsolete after three or four. A '486 is a *minimum* configuration for a CD-R system; a Pentium gives you some breathing room. Pentium II and above is more power than you need, but that's not necessarily a bad thing.
Motherboard: for SCSI, anything with PCI slots is fine. For IDE, anything above UDMA/33 is overkill. See section (5-15) for configuration notes and a warning about certain bus-mastering drivers.
SCSI: the SCSI interface remains a popular choice for CD recorders and CD-ROM drives, though improvements like UDMA/133 are changing the way people build high-end computers. Whether it's built into the motherboard or on a separate card, make sure the host adapter supports ASPI and ASPI for Windows (see section (5-7)). Wide Ultra-SCSI is useful if you're buying a fast hard drive, but CD recorders don't move data fast enough to require it. Bus-mastering SCSI cards are preferred over non-bus-mastering cards, because they can move data to and from system memory directly, without the CPU's involvement, making things faster.
Parallel: some vendors are selling parallel-port CD-R drives. You should have an EPP-enabled parallel port (if you have a Pentium or later, chances are you have one).
Sound: the Creative Labs SB16 and AWE32 boards are widely supported and very popular, but if you're thinking seriously about recording sound through it, you'll want to consider alternatives. See sections (3-12) and (3-13) for other options.
Hard drive: needs to be reasonably fast, and large enough to hold whatever data you plan to put on a CD. IDE hard drives work fine. See section (5-6) for more details.
Video card and monitor: depends on what you want to do. A PCI-based video card is practically a requirement these days, and 17" monitors are inexpensive now. If you're planning on creating multimedia products, scale up.
CD-ROM: SCSI and IDE both work, but some drives work better than others. See section (5-5).
Any Mac of Quadra 700 or higher capability with a reasonably fast disk should be suitable for 2x writing. All PowerMac-class machines, and probably most Mac clones, should work fine at high speeds. PowerBook users should proceed with caution on machines earlier than the 3400 and G3 models.
Any of the SCSI or (for appropriately equipped machines) USB and FireWire recorders should work. Verify with the vendor of the software you plan to use that the drive you have in mind is supported. You may be able to use the internal IDE connector on some Macintoshes as well.
Using the "simulated cut" feature available on Toast and other software is also prudent, at least until you get a feel for the system. Make sure you turn off file sharing before you start a burn, or things will fail if it tries to read a file that's already open. You may also have trouble writing from the boot/system volume, since it will always have files open.
The good news for Mac owners is that the hardware and software configuration for CD-R usually goes rather smoothly.
Besides the obvious question - can it read CD-R discs that you create - there's also the question of how well the drive works as the source device when copying discs. To be more specific:
You can see speed and quality test results on http://come.to/cdspeed.
Older NEC models tend to hog the SCSI bus. Older NEC, Mitsumi, and Acer models (e.g. NEC 3x and Acer 8x) may have trouble reading CD-Rs.
There is one hard and fast rule for direct CD-to-CD duplication: the source drive must be faster than the target drive (e.g. source 4x if target is 2x, source 6x if target is 4x).
A quick summary of features for several models can be found at: http://www.fwb.com/ts/cdt/cdt_support.html
Any recent hard drive will work fine. Back in 1998 this was the subject of some concern, but modern drives are much faster and more intelligent.
There is a fair amount of confusion over what exactly is an "AV drive". A brief discussion is presented here; for more information see Bertel Schmitt's article at http://www.fadden.com/doc/avdrive.txt.
The most important issue is thermal recalibration. Older hard drives would pause for up to half a second (or even up to a full second, depending on who you believe) every so often to adjust the head positioning to the current operating temperature. For most applications this goes unnoticed, but when recording a CD-R you must write the current track to completion without interruption. "AV" drives deal with the problem in a way that doesn't disrupt the disk activity.
A drive that does a quick thermal recalibration is acceptable if the system is otherwise fast enough or the buffer in the CD-R unit or in the recording software is large enough (early drives had only 64KB, while current drives have 2MB or 4MB, making it much less of an issue). You need to be sure that the recorder's write buffer won't empty during the recal period, or you'll end up with a buffer underrun.
Most modern hard drives do smart thermal recalibration. This really isn't something you need to worry about anymore.
What separated a Seagate Barracuda from a Seagate Barracuda AV is that the latter is tuned for AV performance. This was simply a software change that affected cache allocation algorithms, error correction, and other SCSI parameters to get better performance for transfers of large blocks of contiguous data. These sorts of optimizations were very important for digital video running at a few MB/sec, back when that was close to the maximum capability of the drives.
If you think AV optimizations will help you, you should take a look at "Dr. SCSI" at http://www.scsitools.com/.
Some systems have SCSI built in, some don't. This section is intended for PC users who want to add SCSI devices. Owners of SCSI-less Macintoshes should use an interface recommended by Apple.
Using different SCSI adapters for the HD and the CD recorder used to be recommended, but should not be necessary with non-ISA adapters. If your recorder hogs the SCSI bus, though, the HD may not be able to keep the write buffer full. Under some operating systems, particularly OS/2, devices that support SCSI disconnect will work better than those that don't.
In general, the faster the better. PCI or the (now uncommon) VLB is better than ISA, and the board should support (and have enabled) SCSI disconnect. It is not necessary to use Wide or Ultra SCSI for a CD recorder; the speed requirements for all existing recorders are easily met by "narrow" Fast SCSI. If you think you may be buying a speedy SCSI hard drive or other device in the near future, though, you may want to buy a card that supports faster protocols.
You should enable synchronous transfers for devices that support it. Most CD recorders should. If the device doesn't work with it on, turn it off and try again.
The adapter MUST support the ASPI standard (ASPI provides an interface between software and the SCSI controller) for both DOS and Windows.
If you want to boot from a CD-ROM on a SCSI drive, make sure the SCSI card supports booting from removable media.
For some tips on cabling and termination, see Bertel Schmitt's article at http://www.fadden.com/doc/scsi-trm.txt.
The next few sections detail the more popular SCSI cards. There are many others, e.g:
Advansys - http://www.advansys.com/
DTC - http://www.datatechnology.com/
CSC - http://www.corpsys.com/
These are all ISA controllers, good for putting a CD recorder on, not so good for putting a hard drive or fast CD-ROM drive on. If you have an IDE-based system and just want a SCSI card for driving your CD recorder and maybe a scanner or tape drive, any of these (as well as any of the variations of these) will work fine.
The Adaptec 2940 (PCI) is a popular choice -- if not *the* most popular choice -- though some users have reported problems with the Adaptec 2840 (VLB). See the README that comes with Adaptec EZ-SCSI v4.0 and later for some important performance tests you can do with SCSIBench. The 2930 is also a good choice for CD recording.
If you're having trouble writing CD-Rs with the 2940UW, go into the configuration menu (hit Ctrl-A while booting) and make sure the drive is set for 10MB/sec with Wide Negotiation disabled.
A few notes on the 2910, 2920, 2930, and 2940 cards:
The 2940U2W has four connectors (internal 68pin Ultra2-LVD, internal 68pin Ultra2, internal 50-pin, external 68-pin Ultra2) and comes with a special 50-pin cable that ends in a 50-pin (HD) external plate. So you can have both 50-pin and 68-pin external connectors, as well as 50-pin and 68-pin internal connectors. On previous cards, you could only use two connectors at a time, but on this card you can use all five at once.
The ASUS SC-200 is one example of a Symbios Logic 810-based card (in this case, the NCR 53C810). Such cards offer solid performance at a reasonable price, and may be a better choice than the Adaptec cards for many users. (Be sure to examine these types of cards closely though: the least expensive among them are only meant to work with a motherboard BIOS that supports SCSI. This could cause trouble on other motherboards if you wanted to boot from a SCSI hard drive.)
The ASUS SC-875, based on the 53C875 chip, offers Wide SCSI connectors as well.
Symbios Logic is currently owned by LSI Logic. For product information, see http://www.lsilogic.com/products/io_standard/index.html.
Inexpensive SCSI cards based on the LSI Logic SYM53C875 chip. The DC-390U supports Ultra SCSI, while the DC-390F supports Wide Ultra SCSI.
The "SlimSCSI" 1460 and 1480 are PCMCIA SCSI adapters for use in laptops and other portable devices. The 1460 requires a PC card slot and supports SCSI-2, while the 1480 requires a CardBus slot and supports UltraSCSI devices.
The "MiniSCSI" 1350 allows you to connect SCSI devices to your parallel port. If you use this you will be limited to parallel-port speeds, so you may not be able to record at more than 2x.
You can, though there may be reasons not to. The seek times tend to be slower than a standard CD-ROM drive because the head assembly is heavier. Early CD recorders were optimized for writing, which doesn't require fast seeks, and some users experienced jerky video playback as a result. Most current models have pretty good seek times though (about 100ms vs. 80ms for a playback-only drive).
The MTBF on CD-R units has historically been lower than that of CD-ROM drives, so it may be wise to use a different drive for general use to preserve the life of the CD-R. Now that CD recorders are cheap enough to be nearly disposable, though, there's not much point in worrying about them. See also section (5-27) on laser diode lifetime.
(What follows are instructions for getting some of the early consumer CD recorders to work as CD-ROM drives. You shouldn't need to worry about any of this unless you bought an old drive in an auction.)
If you're using Win95, some older CD recorders don't show up as readers without additional drivers, or (for SCSI drives) show up as 8 separate LUNs. (LUNs are Logical UNits, useful for distinguishing between different items loaded in a CD jukebox.) The reason why some older recorders don't show up by default is that they're classified as "type 4" SCSI-2 devices, which is used to indicate write-once devices. Standard CD-ROM drives are "type 5".
HP and Philips used to supply drivers for their older units, and Corel used to supply several drivers for with their CD Creator product. You used to be able to get get a patch from Adaptec at ftp://ftp.adaptec.com/pub/BBS/win95/cdr4up.exe that would allow many type 4 drives, including the Yamaha CDR-100/102 and JVC XR-W2010, to appear as CD-ROM drives, but it appears to be gone. You may be able to find these archived on the web.
If you don't have the drivers, you can still get old SCSI drives to work under Win9X by loading the real-mode drivers like this (example is for an Adaptec 2940):
DEVICEHIGH=C:\SCSI\ASPI8DOS.SYS /DIn Autoexec.bat:
LH C:\WINDOWS\COMMAND\MSCDEX.EXE /D:ASPICD0 /M:12Incidentally, to *remove* the Adaptec cdr4up driver, you should remove the file "CDR4VSD.VXD" from \Windows\System\Iosubsys, and reboot.
For IDE recorders, you need a more specific driver. The manufacturer's web page likely has a link. See also http://www.drivershq.com/ and http://www.windrivers.com/.
This is a general CD-ROM question rather than a CD-R question, but since some of the newer recorders are available in either configuration it seems worthwhile to address it here.
The advantage of a tray is convenience. If you want to put a CD in the drive, you can just drop it in, instead of buying a pile of caddies and hunting for a free one.
The advantage of a caddy is durability. CDs are less likely to be scratched if they're put into a caddy and left there (VERY important if you have children), and the internal mechanism is less likely to collect dust. The tray units usually have a worse MTBF rating, because they have more moving parts.
There have been reports that, at 12x and higher, some CDs will cause loud vibrations in tray models, but work fine in caddy models. Not everyone has had this problem though.
It used to be the case that you had to get a caddy drive if you wanted to mount it sideways, but many tray models have tabs that will hold the CD in place. Having to use the tabs does reduce the convenience normally offered by a tray model though.
Which you should choose depends on your needs and circumstances. If you are planning to write to a disc several times (multisession, packet writing, or anything with CD-RW), you are better off with the disc in a caddy.
With a little extra care, yes. For a Jaz drive, defragmenting the drive right before starting a burn seems to be the key to success. It's also very important to ensure that nothing else is trying to access the drive while the write is underway.
One user reported being able to write at 1x from a DDS tape drive using Seagate's Direct Tape Access, but this isn't recommended. Copying the data to a hard drive and doing the burn from there is much more likely to succeed.
There are no known instances of successful CD-R burns using punched card readers as the source device.
Doing a test run is strongly recommended when using any of these devices.
OPC stands for Optimum Power Control (or Optimum Power Calibration, depending on who you believe; the process is sometimes known as Dynamic Power Control (DPC) or Direct Read During Write (DRDW)). Most CD-R units do a power calibration test before writing to adjust the laser power to the correct strength. Different brands of media and different recording speeds require slightly different power levels. Too much power can create oversized marks which can interfere with each other, and too little power can produce undersized marks which, in extreme instances, can cause read failures.
The recorder reads a recommendation for the initial power level from the Recommended Optimum Recording Power value from the ATIP (section (2-38)) on the disc. This is used as a starting point for a series of write tests in the Power Calibration Area (PCA) of the disc.
Running OPC goes a step farther by actively monitoring the write process and adjusting the laser power as needed. If the writer encounters dust or fingerprints, the laser power can be increased to burn through the obstacles. This is especially useful for discs that are moved around between recording sessions, such as CD-RW discs or multisession CD-Rs.
For more information, see the OSTA white paper on the subject at http://www.osta.org/specs/pdf/opc.pdf. Another good site is http://www.mscience.com/faq64.html.
Audio CD-R/CD-RW recorders are similar to computer CD-Rs, except that they're intended to be part of a recording system rather than attached to a Mac or PC. They have audio inputs and front-panel controls like you'd find on a tape deck. They are usually more expensive than CD-Rs meant for computers. Some CD-Rs have both audio and SCSI-II interfaces.
There are two classes of audio CD-R, consumer and professional. The units targeted at consumers require special audio blanks, and employ SCMS (Serial Copy Management System, section (2-25)) to prevent making copies from a copy. The audio blanks used to be 4x to 5x the cost of computer CD-R blanks and only held 60 minutes of audio, but 74-minute "Consumer Audio" blanks are now available for moderately more than regular CD-R blanks.
The "professional" units use regular CD-R blanks and don't obey SCMS, and generally have a wider set of features and input/output connectors.
If you already have a computer, it's probably cheaper to buy a computer CD-R and a good sound card or digital transfer card (see sections (3-12) and (3-13) for more info). The ability to edit the sound on a computer before writing a CD can be very useful. However, there are some advantages to using an audio CD-R (not all features are present on all models):
You can't copy data CD-ROMs with an audio-only recorder.
(Incidentally, the difference in price for the audio CD-R blanks is due to licensing agreements and volume. The manufacturer pays a royalty to a studio consortium under the assumption that everything recorded to an audio CD-R is pirated material. The technology is identical; the "audio" discs just have a mark that says a royalty has been paid. See also section (7-17).)
It is theoretically possible to convince a "consumer" audio CD recorder to accept regular blanks, but in practice this requires modifying the hardware. Some dealers will sell modified units, with altered firmware or additional circuitry, for a higher price (and perhaps a separate warranty). With the Philips 870/880 units manufactured prior to November 1998, it's possible to trick the recorder by manually ejecting and replacing the disc right before recording. Some of the "code free DVD" sites also sell CD-R chips, e.g. http://www.dvdupgrades.ch/. See also section (7-18).
(And now for some increasingly outdated examples...)
Examples of "consumer" audio CD-R units are the Pioneer PDR-04 and PDR-05 (http://www.pioneerelectronics.com/).
Marantz makes professional-grade CD-R units, e.g. the CDR615 and CDR620.
Philips sells the CDR870 and CDR880 (based on the CDD3600), which support both CD-R and CD-RW media. http://www.acdr.philips.com/products.htm. If you're interested in the Philips CDR765, a consumer-grade dual CD deck, see a detailed article at http://www.gallagher.com/music/cdr.htm and some notes at http://members.tripod.com/~charleswolff/cdr765.html.
HHB sells a "professional" unit, the CDR880. http://www.hhb.co.uk/.
There are many other models and vendors -- Denon, Harmon Kardon, others. Shop around.
In computer terms, hardware is the stuff you can hit with a baseball bat, and software is the stuff you can only swear at. Firmware is software that lives on your hardware. In more concrete terms, the firmware on your CD recorder is what controls the operation of the device, and handles everything from decoding CD-ROM sectors to writing the disc table of contents.
Sometimes there are bugs or missing features that are added by updates. Firmware upgrades have been used to add features like disc-at-once recording and fix bugs like reversed left and right audio channels. Sometimes the upgrade will inadvertently add bugs, causing the recorder to work improperly.
Firmware can be stored in an umodifiable form, such as a ROM chip, or in a rewritable form, such as "flash" ROM. In the former case, firmware upgrades are accomplished by physically removing a chip from inside the device, and replacing it with a new one. Devices with "flashable" firmware, on the other hand, can be upgraded by downloading a new set of firmware over the Internet.
You have to be careful when upgrading the firmware on a drive yourself. If it requires physical replacement, you run the risk of breaking pins off of the chip. Flash upgrades won't result in physical damage, but in some cases a failed upgrade can render the device unusable. Always follow the instructions exactly, and NEVER do an upgrade with anything that didn't come from the manufacturer or a trusted source.
Suppose you want to upgrade your recorder. The first step is to remember famous words of wisdom: if it ain't broke, don't fix it!
The second step is to figure out if your firmware is upgradeable. The manual should tell you. Most drives are, but some exceptions are noted for specific drives in the subsections under (5-1).
The third step is to determine what version of firmware you currently have. Some SCSI cards on PC or UNIX systems will display a list of attached devices when the system boots. There's usually a column with a version number in it.
On a PC running Win95, go into the Device Manager (either from the Control Panels or by asking for Properties on My Computer), and find the CD-ROM drives in the device tree. Select the CD-R drive, hit the "Properties" button, and then click on the "Settings" Tab of the window that opens. Look for "Firmware Revision".
Mac users with Toast can hit Command-R to display the information. If your software doesn't have such a feature, you will need to run SCSI Tools to check the identification string.
The fourth step is to find the upgrade file. Usually the manufacturer's web site will have them. If not, sometimes you can find a repository on the web. (There was a nice one on http://www.ahead.de/en/firmware.htm, but that appears to be gone now.)
The fifth step is to apply the upgrade. This can be trivial or fairly challenging, depending on the device. Be sure to read the instructions *carefully* before applying the upgrade -- if it fails, the recorder could be rendered inoperable.
Section (5-24) discusses the somewhat dangerous practice of flashing a drive with firmware intended for a different drive.
By all accounts, they work just fine. Most such drives are IDE devices with a converter (e.g. an enclosure with a parallel-to-IDE converter).
Parallel-port drives require an ECP/EPP parallel port, which most (all?) machines have. Some BIOSs allow you to switch between ECP/EPP and "standard" mode; if you're having trouble, be sure it's set correctly.
Some people who have bought off-the-shelf parallel-to-IDE converters have found that writing at 4x doesn't work very well. This may account for why all drives that ship with parallel port support are 2x writers.
USB recorders work fine at 4x when connected directly to the computer. You may need to reduce speed to 2x if you use a hub. Some people have reported that their Windows systems were crashing until they turned auto-insert notification off (see section (4-1-1)). Windows users should be running Win98 or later -- Win95b may or may not work. Be warned that some USB SmartMedia readers install drivers that interfere with the ASPI layer; if you have problems with one, uninstall the drivers for the device and run ASPICHK.
You need USB 2.0 to take advantage of drives faster than 6x4x4. Support for USB 2.0 has been spotty, but as of mid-2002 it's becoming more common on new motherboards and software support is improving.
A PC user with USB 2.0 ports discovered that their recorder would only work successfully under WinXP or Win2K. Older versions of Windows wouldn't work.
If you're having problems when disconnecting a device from the USB hub, see http://support.microsoft.com/support/kb/articles/Q253/6/97.ASP.
IEEE 1394 (FireWire/i.Link) devices should only be used with recent versions of Windows on PCs (e.g. Win98SE or Win2K, not Win95, Win98, or WinNT). Linux support for 1394 was still listed as "experimental" in early 2002.
Some personal notes on FireWire:
I bought a Western Digital PCI 1394 card, an ADS Technologies Pyro 1394 Drive Kit, and an HP DVD100i CD/DVD+RW recorder with an IDE interface. As an experiment, I put the HP recorder into the ADS case, and plugged it in.
Under Windows 98SE, I was able to use the drive as a CD-ROM reader and DVD video player. The HP software got a little confused during installation, claiming that it couldn't find the drive, but when asked to record a CD it was able to find the device. However, neither the HP RecordNow software nor Nero was able to successfully record an audio CD. The drive just stopped working a few minutes in.
When the drive was subsequently connected to the IDE bus, it worked fine. Subsequent experiments showed that the problem appears to be some sort of incompatibility with the motherboard -- my VIA-based Soyo K7V Dragon+ seems to be incompatible with 1394 devices. I haven't tried the experiment, but my guess is that the recorder would've worked just fine in the ADS case on a compatible system.
For the curious, http://www.fadden.com/techmisc/my-pcs.htm#1394 has the gory details on what I went through.
(This section assumes you're using a PC.)
You generally want the hard drives and CD-ROM drives on different channels, or CD-ROM accesses can interfere with hard drive accesses. Most older devices can't share the ATA bus, so only one device can be active at a time.
For example, suppose you have a hard drive as master and a CD recorder as slave on the same channel. If you issue a command to write some blocks to the CD recorder, the system can't read anything from the hard drive until the CD write request completes. As long as the system is fast enough, and can read enough data between writes to keep the CD recorder's buffer full, this doesn't create any problems.
If you put the hard drive and the CD recorder on different channels, the commands are allowed to overlap. In practice, on Win9x systems this doesn't make much of a difference, because Win9x won't usually access more than one IDE device at a time. On systems like OS/2 and Linux, the difference is more significant.
Proposals for command overlap (sending commands to multiple devices simultaneously) and command queueing (sending several commands to the same device all at once) were introduced as optional features during the development of the ATA-3 specification. They're part of ATA/ATAPI 4. For command overlap to be effective, both devices on the channel must support the feature. If the hard drive does but the CD recorder doesn't, you won't get much benefit.
If you're not sure that your CD recorder has an ATAPI-4 interface, you probably ought to put it on a separate channel from the hard drive. For information related to this topic, see "Does an old HD or CDROM slow down a new drive?", in section 5.3 of the IDE/Fast-ATA FAQ at http://www.faqs.org/faqs/pc-hardware-faq/enhanced-IDE/part1/.
The recommended configuration looks like this:
Having the CD-ROM drive and the CD recorder on the same channel doesn't necessarily prevent CD-to-CD copying, but you're still better off writing from the hard drive. At high speeds, the CPU utilization for CD-ROM drives without DMA enabled can be very high.
Keep the cables as short as you can. Sometimes the longer (60cm) cables will work fine with one drive but start having integrity problems when two devices are attached.
NOTE: early versions of the Intel PIIX Bus Mastering IDE driver may interfere with the ability to use a CD recorder. The typical symptom is a system hang when writing or test-writing to a disc. The latest version of the Intel driver (which includes an uninstaller) can be found at http://developer.intel.com/design/chipsets/drivers/busmastr/. The Adaptec page http://www.adaptec.com/support/configuration/cdrecide.html also describes the problem.
NOTE: early versions of the VIA Bus Mastering IDE drivers were similarly afflicted. See http://www.via.com.tw/support/faq.htm.
Win95/Win98 users can resolve the bus-mastering IDE driver problems by installing Win98 Second Edition (a/k/a Win98SE) after removing any manufacturer-supplied bus-mastering drivers.
The ASPI (Advanced SCSI Programmer's Interface) layer is used during CD recording, even for IDE recorders. See section (4-44) for information on how to make sure you have what you need. The original Win95A/B WinASPI may have problems with IDE recorders.
Yes. If your drive is in PIO mode you can get bad results when recording or ripping audio. (PIO is "Programmed Input/Output". The computer has to sit and wait for small I/O operations to complete, instead of handing the drive a big data buffer and letting the drive manage things. DMA, "Direct Memory Access", is much more efficient than PIO.)
Sometimes Win2K and WinXP will revert to PIO mode when a number of DMA errors are detected. See http://www.microsoft.com/hwdev/tech/storage/IDE-DMA.asp and http://www.gmayor.com/cd_writer_udma_mode.htm for details.
Under Win98, you can toggle the DMA setting by opening the Control Panel window, double-clicking on System, selecting the "System Properties" tab, expanding the "CDROM" branch, selecting the device, clicking on "Properties", clicking on "Settings", and then checking or unchecking the "DMA" checkbox. Win2K, WinXP, and later versions of Windows have the setting in a similar location. Under Win2K, you can set DMA on a per-channel basis. Under WinXP, select the adapter that the drive is on rather than the drive.
On 1990s-era hardware and software, the answer was more of a "maybe". The rest of this section is "historical".
Some drives in some configurations will not work correctly, so the right answer is "try it and see". If you are having lots of problems getting a drive to work, turn it off. If you're running with it off, and are having performance problems, turn it on.
As with any other "try and see" procedure, don't change more than one thing at a time. For example, don't rearrange your drives and toggle DMA without doing some testing in between. Otherwise, if something breaks, you won't know which change caused it.
This was an interesting question back in the early part of 2001, when CD-RW support was not present in all drives. All CD recorders made today support both CD-R and CD-RW media. However, the question is still of some academic interest, so the original answer follows.
It depends on what you're doing. CD-R media is incredibly cheap these days, so using CD-RW to burn a a test disc doesn't make much sense unless you're burning a *lot* of test discs. Besides, CD-RW discs aren't readable on many older CD-ROM and audio CD players.
The manual for Easy CD Creator Deluxe v3 says that CD-RW discs are "more cost effective for near-line data storage requirements than CD-R." The definition of near-line storage puts it somewhere between online storage and offline storage.
On the other hand, if you're expecting to use packet writing to treat the disc as a big floppy, it may be useful. You should consider other forms of media for such purposes though, such as Jaz drives, which are faster and hold more, but are slightly harder to find readers for (but only slightly: CD-RW discs aren't readable on all drives, and packet-written discs may not be readable under some operating systems).
Software developers who need to create test CDs frequently will find CD-RW invaluable.
Historically, each manufacturer of CD recorders used a different command set, and perhaps even altered the commands with each new recorder. This has placed a significant burden on CD-R software authors, who have to write new drivers for each new device.
MMC (Multi Media Command) compliant recorders use a common command set. Programs that can write to one MMC-compliant recorder should be able to write to all others, and consumers should be able to use their choice of software without the long delays usually associated with the introduction of new hardware.
The reality is not so kind, unfortunately, due to firmware bugs or deliberate deviations from the standard. Do not assume that a particular piece of software will work with your recorder simply because it works for other MMC-compliant devices.
The spec sheets for recorders usually indicate whether or not the drive is MMC compliant.
The MMC-2 standard is documented in ANSI/NCITS 333-2000. You can buy a copy of the standard from http://www.ncits.org/ (specifically, http://www.techstreet.com/cgi-bin/detail?product_id=223931).
The choice of what hardware to buy is dictated by software availability. Find the software you want to use (common choices include "cdrecord", listed in section (6-1-20), GEAR in section (6-1-3), and CDR Publisher in section (6-1-9)). All support a variety of recorders, primarily SCSI devices.
Consult the software manufacturer's web site for any specific recommendations.
It's possible to get IDE recorders working under Linux, by installing an "ide-scsi" module that makes the recorder work more or less like a SCSI device. This is similar to what the Windows ASPI layer does for IDE devices. See the CD-Writing HOWTO for more details (try http://www.linuxdoc.org/HOWTO/CD-Writing-HOWTO.html).
The Sun CD FAQ at http://www.datamodl.demon.co.uk/suncd/ has some helpful tips on using CD recorders and creating bootable CD-ROMs for Solaris machines.
You need a way to connect the recorder to the laptop. After that, it's really no different from a desktop.
You can connect a typical recorder via USB (if you have a USB connector), SCSI (if you have a port or want to buy a PCMCIA SCSI card like the Adaptec 1460), FireWire (if supported or you have a PCMCIA 1394 card), or parallel port. SCSI is the fastest, but PCMCIA SCSI adapters tend to be expensive. FireWire is the next best bet. USB is a good choice, and should be available on most recent laptops, but you're limited to recording at 6x or less with USB v1.x.. Parallel port works fine, but you will probably be limited to recording at 2x.
A small selection of portable CD recorders is now available. These are small, battery-powered devices that come with a PCMCIA connection. Examples include the Ricoh MP-8040SE and Smart & Friendly Pocket RW.
In some cases it may be possible to replace the CD-ROM drive included in the laptop with a CD recorder.
Search section (5-1) for "portable" devices.
If the software options described in section (3-17) are insufficient, you may want to buy dedicated hardware. You can learn about the types of equipment available at http://www.octave.com/library/cdduplicating.html.
The purchase of a CD recorder often results in what used to be an unusual situation: a machine with two CD-ROM drives in it. This leads to a number of interesting phenomena, usually having to do with poorly-written software that can't figure out which CD-ROM drive it's supposed to use.
CD-ROM drives are typically connected to a sound card via a small cable (a couple of wires twisted together, ending in small molex connectors). This allows audio CDs to be placed in the CD-ROM drive and played through the speakers attached to the sound card. Some people, upon discovering that they have two CD-ROM drives and can use both simultaneously, want to connect both drives to the sound card's input.
This is where the trouble starts. Sound cards often only have one input. The immediate temptation is to buy or construct a Y-cable, but this won't always work. The trouble is that Y-cables only work when you have a single signal and more than one listener, like a stereo that sends its output to two sets of headphones. The situation with two CD-ROM drives is of two outputs and one listener.
Connecting two outputs together is, in general, a bad idea. Remember that electricity isn't like water: it does not come out of the output and flow downhill. The voltage at any point on the wire (ignoring minor distortions) is going to be exactly the same. So if you have a device that's trying to set it to one level, and another device that's trying to set it to another level, the two devices are going to fight, and the results aren't going to be what you want.
In some cases, if a device is inactive, it will allow its output to "float". The other device can set the voltage to whatever level it wants. So long as you only use one device at a time, all is well. Many devices, however, force the output to ground level when not in use. This generally manifests as a volume level that is almost inaudibly quiet.
Devices that combine multiple audio inputs into something reasonable are called "mixers". Buying one and embedding it into your PC case is probably not the best solution.
One possible option, if you're handy with the soldering iron, is to rig up a mechanical switch that selects which signal gets passed to the sound card. So long as you weren't planning to play two audio CDs simultaneously, this should work well.
Some sound cards have multiple connectors on them, suggesting that the card itself could handle multiple inputs. More often than not, these connectors are not electrically isolated, so even though they're not sharing the same cable they will still cause the devices to compete. If the sound card isn't advertised as allowing multiple independent inputs, don't assume it can.
Some of the Sound Blaster cards, e.g. SB Live!, do have two independent inputs ("CD in" and "AUX"). Stay away from the TAD (Telephone Answering Device) connector though, it's monaural. You may need to un-mute the auxiliary input in the volume control panel.
You can get an inexpensive Y-cable with a "passive mixer" from "Cables N Mor" at http://cablesnmor.com/cdrom.html. If you're the build-it-yourself type, some instructions for building a similar cable can be found on http://www.math.wisc.edu/~wilson/miscellany/mixer.htm.
A player spinning a CD at 1x reads 75 sectors per second. On a CD-ROM, where a sector has 2048 bytes, this is exactly 150KB/sec. On an audio CD, with 2352 bytes per sector, this works out to about 172.27KB/sec. (Note for the nit-pickers: the actual bit rate is considerably higher, because of EFM, CIRC, L2 ECC, and other magic acronyms. The channel bit rate is 4.3218MHz. See Ken Pohlmann's _Principles of Digital Audio_, 4th edition, page 249.)
In terms of revolutions per minute, the answer varies depending on which part of the CD is being read. At 1x, the speed at which bits flow under the read head (the "linear velocity") needs to be fairly constant. You can get more bits in a circle at the outside of the disc than you can in a circle at the inside of the disc, because the circumference is greater. This means that the disc needs to spin more slowly (reduced "angular velocity") at the outside than it does at the inside.
To play an audio CD, you always want to be reading at 1x. This means you need a constant linear velocity that gives you 172.27KB/sec. The angular velocity changes as you move toward the outside of the disc.
To read files from a CD-ROM, you want to be reading as fast as you can. This means you'd like to maintain a constant angular velocity, spinning as fast as the spindle can go, with a linear velocity that increases as you move out to the outside of the disc. This is why a drive like the Plextor 12/20 reads at 12x at the start of the disc and 20x near the end.
In practice, there is a maximum angular velocity because of physical constraints, and a maximum linear velocity because of hardware and software constraints. This results in drives that use constant angular velocity for the first part of the disc, but limit themselves to a maximum linear velocity. As the read head moves further out on the disc, the drive switches to constant linear velocity mode.
Devices that always spin at the same rate are called CAV (Constant Angular Velocity) drives. Devices that maintain a fixed linear velocity are called CLV (Constant Linear Velocity) drives. Devices that switch from CAV to CLV when the maximum speed is reached are called PCAV (Partial Constant Angular Velocity) drives. Most of the recent high-speed CD-ROM drives are PCAV. Devices that are CLV, but use different speeds on different parts ("zones") of the disc, are called ZCLV. Most CD recorders use CLV while writing, but some (e.g. 20x and higher) use PCAV or ZCLV.
See http://www.plextor.be/english/technical/zoneclv.html for a graph illustrating ZCLV. http://www.cdspeed2000.com/go.php3?link=faq_general.html has some nice charts showing CDSpeed output on different drive types.
You can compute how long it will take to record a disc with a CLV drive by taking the amount of data and dividing it by the record speed of the drive. A 74-minute disc will take about 19 minutes to record at 4x and a little under 10 minutes at 8x. With a PCAV drive, this calculation is no longer valid, because the velocity changes as the write head moves outward.
In terms of actual rotational speeds, a disc being read at 1x spins at about 530rpm when reading near the center of the disc, slowing to about 200rpm at the outer edge. The linear velocity is constant, ranging from 1.2 m/s to 1.4 m/s depending on the disc. Discs with longer playing times (e.g. 74 minute discs vs 60 minute discs) use the slower velocity.
It has been stated that, at a rotational speed equivalent to about 50x at the inside of a disc, the polycarbonate starts to deform and the disc becomes unreadable. Experiments (e.g. an episode of the "Mythbusters" TV show from 2003) have demonstrated that discs will warp when they get up around 25,000 to 30,000 RPMs. However, recent 52x drives only read data that quickly from the outside of the disc, actually reading at about 21x near the inside. This requires a speed of 10,000 to 12,000 RPM, which is safe for discs in good condition. Reading at 52x from the very inside of the disc would require a speed of about 27,500 RPM, and read data at 137x near the outside.
Discs with minor defects can and will shatter at these speeds, so some care must be taken with drives rated at 40x and above. See section (7-25) for more information.
An unbalanced disc can cause noisy vibrations in high speed drives. Some devices will actually reduce the spindle speed if the vibrations become too severe.
Incidentally, "1x" on a DVD-ROM drive is 1353KB/sec, which is roughly 9x the speed of a "1x" CD-ROM drive. A 16x DVD-ROM drive reads at a speed equivalent to a 144x CD-ROM drive! The DVD doesn't actually spin 9x as fast, though, because the DVD "bit density" is higher. The drive can read roughly three times more data in a single revolution from a DVD than it can from a CD. (Incidentally, the 1353KB/sec figure comes from the DVD maximum user data rate of 11.08Mbps, where the 'M' is 1000*1000.) For more details, see http://www.dvddemystified.com/dvdfaq.html#4.2.
Generally speaking, no, though warnings have started to appear.
One proposed line of reasoning is that the lower reflectivity of CD-R media causes the laser to work too hard. This only makes sense for players with an AGC (Automatic Gain Control) circuit, in which the laser power adjusts automatically. This feature is generally found in newer players, because it's required for reliable playback of CD-RW discs.
It seems unlikely that a player with an AGC would fry itself while running at a valid power level, unless the device were poorly constructed. In any event, the reflectivity of CD-R is close to that of CD -- if it weren't, CD-R would have the same playback compatibility problems that CD-RW has.
The laser shouldn't have to work any harder to read CD-R. It's possible that some devices might "strain themselves" over CD-RW discs, but any device built to work with CD-RW should be able to handle the media without self-destructing.
A more likely scenario relates to differences in physical dimensions. One car dealer claimed that CD-R media is too thin, causing their 6-disc changer to occasionally grab two discs and jam itself. On the opposite end of the spectrum, some "slot-in" dashboard players will get stuck ejecting a CD-R that has had an adhesive label added, because the disc is too thick.
It's possible that the players with warnings simply don't support CD-R well for one reason or another. Rather than admit to poor construction, the manufacturers are trying to make it seem like there's something wrong with CD-R media.
Not in the sense that you can make a slow drive work faster, but in some cases you can make a speed-limited drive work at its full capacity.
Every drive model in a manufacturer's lineup costs additional money to make, because the manufacturing line has to maintain a larger inventory of parts and has to re-tool the assembly line whenever they switch production. In the world of high-volume, low-margin products, eliminating these costs can be a huge win. Changing hardware components also creates opportunities for things to fail, so every new hardware design must be extensively tested. (The above is true of many consumer electronics products, not just CD recorders.)
Some manufacturers build a high-speed drive and then use firmware to limit the drive to slower speeds. There can be technical reasons for doing this -- it's possible the parts they're using don't work well or they haven't finished getting the firmware working well at higher speeds -- but often its for marketing reasons. The higher-speed drives can initially be sold at a higher cost. If you build a 20x-capable drive, you can sell it for more than the same drive limited to 12x performance. By selling the same drive as the 12x unit and the 20x unit, you're cutting manufacturing costs even if the 20x-capable parts cost slightly more.
(CPU manufacturers typically build chips for a single speed and then sort them into speed bins based on how quickly they were able to run before they got flaky. The expensive "turbo" versions of your favorite graphics card are the same hardware as the base versions, but they ran at a faster speed without crashing. You're paying a premium for the performance boost, but it follows the laws of supply and demand: the chips that run at the highest speeds have the lowest yields, hence they cost more.)
Computer overclockers like to push the boundaries of what their components can do by assuming that the chip manufacturers put some tolerances into the bin-sorting, meaning that they can run the chip faster than rated without it becoming unstable. Or at least not *too* unstable.
With CD recorders, the speed differences might be due to hardware limitations or might be due strictly to marketing reasons. The common experience among "overclockers" is that the firmware change simply converts the drive from one kind to another. It's unclear, however, if such updates introduce more subtle problems, such as worsening the jitter present in audio recordings.
It should be pointed out that updating your drive with firmware for a different drive is VERY DANGEROUS and could result in your drive being unreliable or irrevocably dead. You should not attempt to "overclock" your recorder unless you were planning to get rid of it anyway.
Remember, this change only works on drives that were deliberately underpowered, so for many devices "overclocking" simply isn't an option. For details on performing these modifications to a variety of drives, see:
When in doubt, read the manual. If a tech support phone number is included, call it. Read section (5-15) for information on IDE configuration for a PC.
Yamaha has some interactive instructions for the PC on their site at http://www.yamaha-it.de/england/firststeps-english/index.htm. (The navigation is a little counter-intuitive, but it's okay once you get started.)
The book _CD and DVD Recording for Dummies_ by Mark L. Chambers has a section on drive selection, installation, and troubleshooting. If you're new to CD recording, the software tutorials may be helpful as well.
About the same as a CD-ROM drive, even when recording. Some simple experiments suggest that the only significant power drain occurs when the disc is spinning up. Some personal notes follow.
I connected an external Plexwriter 8/20 through a "Watts Up?" power meter. The Watts Up? device is designed for moderate loads (20W up to about 1700W) and isn't good at detecting small fluctuations, but it's accurate enough for this purpose. I connected the CD recorder and a fan drawing 50W through the meter, and subtracted 50W from the results.
When completely idle, the CD recorder and its power supply draw 8-9W. Since the recorder isn't actually doing anything, I'm guessing most of this is loss in the power supply itself. In any event, it establishes an idle-load baseline.
While playing an audio CD through the front panel headphone jack at 1X, there was no change in power usage.
While playing an audio CD through Windows Media Player, the load increased to 9-10W. I got a similar drain while extracting audio at 8x with jitter correction and at 20x without jitter correction (about 13x actual speed, according to Nero). Recording a disc at 8x gave the same result.
The only time I saw the recorder draw more than 10W (1-2W above idle) was during transitions. Inserting an audio CD gave a quick 16W pulse, and there were similar small blips at the start and end of recording the CD. Spinning up the spindle appears to draw an extra 6-7W over the idle load, but very briefly.
A drive with a higher speed rating would draw more power while spinning up, but would probably use the same amount while actually doing work. While installing Linux on a different system with an Asus 52x CD-ROM drive, I noticed the load for the entire system went from around 50W when idle to a fairly stable 90W while doing CD media verification. How much of that was the drive and how much was the CPU is unclear -- the load on the system would go from 50W to 70W when quickly raising and lowering a window under X11 -- but it's clear that there's more to the story than the drive itself.
My earlier hypothesis -- that CD recorders draw significantly more power when recording -- appears to be incorrect. There have been cases where people could do test writes but not actual writes, and solved the problem by upgrading their power supply. However, this appears to have more to do with the power supply's stability than changing load requirements. The power supply that fixed the problem may have been more reliable, or perhaps the old one was always overtaxed and the problem didn't manifest itself until something requiring precise power management was in use.
Yes, eventually. Depending on a number of factors, though, it's quite possible that your device will suffer mechanical breakdown or simply become obsolete before that happens.
There are many different ways to construct a laser diode. Different approaches result in different wavelengths, maximum power levels, and lifetimes. The lifetime of a laser is usually measured as MTTF (Mean Time To Failure) at a particular power level and ambient temperature (e.g. 10,000 hours at 5mW and 50 degrees Celsius).
Higher power levels mean higher heat dissipation -- the optical conversion efficiency of a laser diode is around 30% -- and in the semiconductor world, more heat usually equates to shorter lifetime. Recording for an hour at high speed will take a greater toll on the laser than playing a CD for an hour.
The bottom line is, there really isn't anything you can do to make the laser last longer. It'll last a very long time when used to read CDs, so there's no point in reserving the drive just for recording. It might last a little longer if you use lower recording speeds, but if you're willing to do that then why pay for a high-speed recorder?
Sony Semiconductor's "Laser Diode Guide" is available from http://www.sony.net/Products/SC-HP/tec/catalog/laser.html.
Laser diodes can suffer catastrophic failure (they suddenly stop working) or gradual degradation (reduced optical power for a given input power level). The power calibration sequence ( (4-13), (5-11)) automatically adjusts the power supplied to the laser for a given disc and write speed, so reductions in output are compensated for automatically. However, if the laser's efficiency is reduced, more DC power must be supplied, more heat is generated, and the likelihood of failure increases.
For a discussion of laser diode reliability, see http://www.bostonlaserinc.com/art1.php.
Top of page
FAQ Copyright © 2010 by Andy McFadden. All Rights Reserved.