Reviews Leaderboard Database Reference Search StorageReview Discussion Reliability Survey Search About StorageReview.com Contents

Yamaha CRW2100S

  February 20, 2001 Author: Tim Zakharov  
Special thanks to Hypermicro.com* for providing the evaluation unit.

* Remember, mention StorageReview.com when ordering from HyperMicro and receive FREE shipping!


Introduction

Back on January 17th of this year, we examined Yamaha's CRW2100E ATAPI burner. Yamaha's offering was the first 16X writer to reach retail channels. While its 16X write speeds were indeed faster than the best 12X burners, there were a number of issues with the drive that kept us from recommending it over burners like Plextor's PX-W1210TA. Since then, TEAC's CD-W512E, another 12/10/32 burner, has edged out Plextor's outstanding drive for the top spot on our Leaderboard, leaving Yamaha's next-generation offering lagging behind its predecessors.

Recently, Yamaha followed up the CRW2100E with an identically spec'd SCSI model. The CRW2100S follows in the footsteps of its ATAPI brother, becoming the first 16 X SCSI burner to reach store shelves. Does it improve upon the CRW2100E's performance? Have issues that plagued the ATAPI model been resolved? Let's find out.

Specifications

Specifications according to Yamaha's drive manual:

  • Maximum write speeds of 12X-16X Partial CAV
  • Maximum rewrite speeds of 8X CLV
  • Packet-writing speeds of 4X-10X CAV
  • Read speeds of 40X Max (CAV)
  • Digital audio extraction speeds of 40X Max (CAV)
  • 8 MB buffer
  • 160 ms average random access time
  • SCSI-3 (Ultra SCSI) interface

For more information, click here to see Yamaha's online product page. Note that on this page, Yamaha makes no mention of partial CAV or the special circumstances required to rewrite at 10X.

Hyper Microsystems sent us the retail box, allowing us to see firsthand what Yamaha includes with their drive. First off, we're very impressed with the plastic shell used to secure the drive in the box. It appears to provide better shock protection than the typical foam drive holder. Also, the manual is extremely thorough and user-friendly. The rest of the contents:

  • 50-pin SCSI cable
  • Standard 3-pin audio cable, mounting screws and washers
  • One blank Yamaha 650 MB 16X CD-R (no CD-RW disc is included)
  • Easy CD Creator/DirectCD burning software
  • Adobe Acrobat software
  • Adobe PhotoDeluxe Business Edition v1.1
  • Adobe PageMill 3.0
  • Adobe ActiveShare
  • 39 free MP3 songs (various unknown artists)
  • Music Match Jukebox software
  • Neato CD labeler, Media Face II software, and a few blank labels to play with
  • Leaflet describing how to download free "Fast Audio Rip 1.0" software
  • Registration card

This is probably the most complete retail package we've encountered.

The rest of the drive's vitals: Unlike the CRW2100E, the SCSI version does not feature a digital audio-out connector. There is a single LED on the front that glows amber when there is no disc in the drive and cycles between different combinations of solid and blinking amber and green lights, depending on the drive's activity. The manual includes a chart listing all the possible permutations.

The drive arrived with Yamaha's most recent firmware revision, 1.0h, already installed.

While the CRW2100S never exceeded lukewarm temperatures in our stress tests, noise levels were extremely high during full RPM reads. When we say "noise," we also mean vibration - this drive (as well as the ATAPI version we recently reviewed) is easily the loudest optical drive we've ever tested.

Current online pricing for the retail CRW2100S kit is about $275, about $50 higher than the CRW2100E.

For an overview on methodology, click here.

CD-ROM Performance Results

Low-Level Measurements

[an error occurred while processing the directive]

[an error occurred while processing the directive]

CD Winbench 99 transfer rate measurements peg the Yamaha's sequential transfers at 18X on the innermost tracks and at 37X on the outer edge of our test CD. The CRW2100S comes in a hair slower than its ATAPI brother here. Both are under Yamaha's 40X spec, but as we learned in our CRW2100E review, the Yamaha does not reach its spec until the outer edge of a completely filled 700 MB disc. The transfer rate graph reveals no problem areas, save for a small dip at around 300 MB. This is normal... most drives we've tested exhibit such minor dips.

In CD Winbench 99's access time test, the CRW2100S yields a 137 ms average. That's a bit quicker than Yamaha's 160 ms spec, but 7 ms slower than the CRW2100E. This, though keeps the SCSI Yamaha slightly ahead of the Plextors. We're pleased to see most CD-RW access times improving over time.

CD-ROM Winmark 99

[an error occurred while processing the directive]

CD Winbench 99's CD-ROM Winmark test runs through a timed script of routines ripped from a variety of popular software programs. The score presented here is an average drawn from four different test discs.

The CRW2100S comes in at 1002 KB/sec, 12% slower than the ATAPI Yamaha. This places the SCSI Yamaha last in our comparison. It appears like the slower measured access times contribute to this 12% difference. In addition, both Yamahas exhibit lower-than-expected scores compared to the competition. With 40X reads and access times in the 130 ms range, Yamaha Winmark scores should be significantly higher. The Yamahas are prone to spinning down during this test (it happens with both Yamahas and with all four of our test discs)... most likely a cause of the low scores. We'll revisit this shortly.

File and Disc Copy

[an error occurred while processing the directive]

Our file copy test emphasizes sequential transfer rates by copying a single 634 MB file. The Yamaha's 40X read specs help it garner times significantly lower than the 32X competition from Plextor, Sony and Ricoh. The CRW2100S clocks in a hair faster than the ATAPI Yamaha.

The disc copy test introduces random accesses due to the multiple files and folders on our test CD. The CRW2100S finishes a bit slower than the CRW2100E... what one would expect from the low-level access time results. We're not certain why the Yamahas lag in these tests; based on their low-levels, they should be near the top of the list rather than the bottom. What we do know is that both Yamahas spin down a lot during this test, undoubtedly slowing copy times.

Combined with the Winmark scores presented above, these results raise questions on the CRW2100 series' ability to read at full speed during random accesses. In two separate benchmarks, with two separate sets of discs, we've encountered the same random-access problems.

Digital Audio Extraction

[an error occurred while processing the directive]

The ATA CRW2100E had problems extracting smoothly from our test CD-DA. The SCSI CRW2100S has very similar problems. Brand-new discs extract at or close to Yamaha's 40X Max spec, while even slightly scratched discs lead to inconsistent DAE speeds - lots of spinning down mid-extraction and dips and valleys in the DAE graph. CDSpeed99 measures minimum extraction rates of just under 17X while average DAE across the disc is about 25X. Because of all the spinning down at the outer edge of our CD-DA, we could not measure consistent DAE speeds towards the end of the disc. The presented DAE graph is the one most typical out of the various runs performed.

CDDAE99 is an audio ripping program that doubles as a benchmark. The program extracts audio tracks to .wav files one at a time or the whole disc at once. CDDAE99 extracts each track twice; the second verifies the original (verification can be disabled for faster rips). Differences are noted as errors. Tests with CDDAE99 show an average extraction rate of about 24X across the entire disc. Again, the drive spins down when faced with any sort of scratched CDs. This typically occurs near the outer region of the disc. Fortunately, no errors were detected, either in CDDAE99 or in CDSpeed99: We listened to selected .wav files extracted with CDDAE99 and found no audible defects. Thus, it appears that extraction quality is not affected by the spindowns...only speed. Even so, the CRW2100S averages 24X-25X - slightly faster than comparison drives.

CD-R Media Compatibility and Performance

[an error occurred while processing the directive]

Like the CRW2100E, the SCSI Yamaha displays excellent sequential read performance with CD-R and -RW media. Read speeds with TDK 12X CD-Rs top out at 38X, equal to pressed CD read speeds. Speeds remain in the same range with Verbatim 16X and Imation 12X media. The Imation is a 700 MB disc filled to capacity. The Yamaha can read the outer edge of this disc at 40X - fulfilling manufacturer claims that this drive features 40X Max reads.

CD-RW Media Compatibility and Performance

[an error occurred while processing the directive]

Yamaha's CRW2100 series continues to be one of the fastest CD-RW readers around. The SCSI unit reaches nearly 39X on the outer edge of our test disc - slightly faster than it reads pressed CDs. We measure similar performance with Verbatim, Memorex, and Sony CD-RWs.

However, when testing with our high speed TDK CD-RW, CDSpeed99 errors out every time. To find the source of this error, we first created a 2nd copy of our test disc on a physically different TDK high speed CD-RW. The error persisted. Next, we used Explorer to copy the contents of these discs to the hard drive. We had no problems copying the contents of both discs to hard disk, though the spin downs we previously noted continued. In fact, copy times were a bit slower than the pressed CD results presented in our File and Disc Copy section above. Our best guess is that CDSpeed99 forces the drive to read at the fastest speed possible, whereas actual copies through Windows Explorer allows the Yamaha's firmware to slow reads down if impending read errors are detected. Other possibilities, of course, are bugs and compatibility issues with CDSpeed99.

Write/ReWrite Tests

CD-R Based Duplication

[an error occurred while processing the directive]

When duplicating our audio CD, the CRW2100S takes quite a bit longer to image the disc than the CRW2100E. However, burn times are much closer, with the CRW2100S completing the process in 5:29 vs. the CRW2100E's 5:25. Total duplication time puts the CRW2100S nearly a minute behind its ATAPI sibling, due to the slower imaging.

Our data CD takes 2:44 to image and 5:39 to burn. These results are about equal to the ATAPI Yamaha, keeping both CRW2100 drives significantly ahead of the 12X pack in total duplication time.

CD-RW Based Duplication

[an error occurred while processing the directive]

Like the CRW2100E, the SCSI Yamaha rewrites at 8X speeds during normal burns. Both Yamahas duplicate our test disc to CD-RW in 12:28.5. Only Sony's 8X rewriter is slower in our comparison. The other drives rewrite at 10X, keeping them well ahead.

CD-R Based Stress Tests

We stress our optical drives by attempting a scripted burn while the testbed runs certain programs. In what we've coined our moderate stress test, this test runs while the testbed cycles through Unreal Tournament's introductory flyby sequence. Every drive tested so far completes this burn at its maximum write speed. The CRW2100S is no exception, completing the task easily at 16X. This reinforces the fact that the majority of today's burners, regardless of interface, are capable of peak performance even under fairly heavy loads. Our testbed's CPU is decidedly low-end at 450 Mhz. Faster systems mated with a modern burner likely would complete full speed burns at even higher system loads.

Speaking of high system loads, our other stress test is a worst-case scenario. Most users don't put this much strain on their system while attempting to complete a burn, but we like to test the limits of performance at SR. We attempt the same scripted burn while Winbench 99's CPUmark99 test executes. During this test, the CPU and CPU cache are completely loaded, leaving few resources free for anything else. So far, SR has reviewed nine burners. The fastest that any drive without BURN-Proof or JustLink could complete this test was 4X. Those drives with buffer underrun prevention technology can complete a burn at maximum-rated speeds under these conditions, but with pauses factored in while the drive waits for its buffer to refill, speeds typically end up just below 5X.

So how does the CRW2100S do? Without any buffer underrun technology, the drive tops out at 4X despite its 8 MB buffer. In other words, it performs no better than drives with 4 MB buffer, such as Sony's CRX160E. Keep in mind; even at speeds as slow as 4X, an 8 MB buffer only takes about 13 seconds to empty.

[an error occurred while processing the directive]

DirectCD Formatting

The CRW2100S completes a full DirectCD format in 13:28. This is 11 seconds faster than the CRW2100E, but slightly behind TEAC's CD-W512E. The SCSI Yamaha quick formats a high speed CD-RW in about 34 seconds.

DirectCD Packet-Writing Performance

Like the CRW2100E, the SCSI Yamaha rewrites at 4X-10X CAV during packet-writing. As such, the CRW2100S takes nearly five minutes to copy our 195 MB folder via Windows Explorer. This is slightly slower than the ATAPI Yamaha. Even Sony's 8X CLV speeds yield much faster times.

DirectCD CD-RW Erasing

Unfortunately, the CRW2100S doesn't work with Adaptec's CD-RW Eraser program. We tried various brands and speeds of CD-RWs, but CD-RW Eraser loads up with the "erase" button grayed out every time. We swapped a different burner (Plextor's 16/10/40) into the testbed to erase the discs we'd formatted to DirectCD on the Yamaha. The Plextor has no issues with CD-RW Eraser. In fact, Yamaha's CRW2100E has no problems either when erasing CD-RWs under identical conditions. Since both Yamahas were tested with the same firmware (1.0h) and same software, its unclear why this occurrs. Could there be a difference between the CRW2100E's 1.0h and the CRW2100S's 1.0h?

Conclusion

Yamaha's SCSI entry in the high-speed CD-RW market has a lot going for it. First off, it's 16X and SCSI. Second, it claims 40X Max read speeds. Third, it claims 40X Max audio extraction. Finally, it advertises 10X rewrite speeds. All are either milestones or industry standards in recording/rewriting technology. Unfortunately, there are issues in all areas.

First, Yamaha chose partial CAV writes instead of full 16X CLV writes as competitors have debuted. In practice, this results in a paltry 15 second difference when burning 650 MB; the implication, however, is that the consumer gets less than what's advertised. Write and rewrite speeds up to this point have always been CLV - when one saw "12X" on the package, 12X was what one got. Yamaha puts "16X" in bold print on the front of their box, leaving the details in the fine print. If the other manufacturers took the P-CAV route because of some technical limitation inherent to achieving such speeds, this would be a non-issue. But 16X CLV drives from Plextor and TDK have already arrived with more on the way.

Next, we could not achieve 40X read speeds on pressed media. Only when we test with a custom-made 700 MB CD-R can we reach 40X at the outer edge. The drive also spins down frequently in operations involving random accesses. In some cases this brings performance levels down below what we've measured from competitors' 32X offerings.

More importantly, the drive exhibits such high noise levels when reading at full speed that we could not tolerate it. An example of how loud this drive is: the testbed case resides on a carpeted floor an inch away from the side of a wooden desk. When the Yamaha operates at maximum rotation speed, vibrations carry from the drive through the testbed case, through the carpet, to the desk, and can be felt anywhere along the desk. In fact, vibrations can easily be felt while using a mouse and keyboard on the desk. As for noise levels, when the drive is set up to run a sequential transfer rate test, it can be heard from the floor below. It's that loud.

40X Max audio extraction can be obtained only on scratch-free discs. Even slightly scratched discs that don't faze other drives cause the Yamaha to consistently spin down when reaching the outer portion of the CD-DA. The end result: average extraction speeds are only slightly faster than 32X Max DAE from the competition.

Finally, the Yamaha woefully underperforms when writing to CD-RW media. Claimed 10X speeds occur only at the outer edge of a packet-written disc. Standard rewriting occurs at 8X CLV, while packet-writing occurs at 4X-10X CAV. We found Sony's 8X CLV packet-writing to be faster.

Although the Yamaha does outperform previous-generation 12/10/32 burners in a couple of areas (most notably while writing at 16X), in most cases its performance is either a few percentage points better, equal, or worse than the competition.

So to whom do we recommend this drive? We feel comfortable recommending the CRW2100S only to those who need a 16X burner with a SCSI interface right now, and those who will only use the drive to burn CD-Rs at 16X. Those who want to use their burner as a reader as well as those who will be doing any amount of rewriting or packet-writing are better off considering 16/10/40 burners from the competition.


HOME | ARTICLES | LEADERBOARD | PERFORMANCE DATABASE | REFERENCE GUIDE
COMMUNITY | RELIABILITY SURVEY | SUPPORT SR! | ABOUT SR |

Copyright © 1998-2005 StorageReview.com, Inc. All rights reserved.
Write: Webmaster