Doogie Posted February 14, 2008 Share Posted February 14, 2008 I almost hate to post this, but I recently received a 16GB Voyager GT and it its not performing well? I've used ATTO Benchmark, HD Tune and HD Tach on 3 different platforms and none of them showed read values higher than 17MB/s? Worse yet, the write speeds haven't topped 7MB/s! It may be a bad drive? In real world testing with Migo, it doesn't perform as well as a recently purchased 8GB Voyager (non-GT)! I use Migo to transport about 4GB of information between work and home. If you're not famillar with Migo, think of it as transporting your desktop, user files and email from computer to computer...you actually work from the drive when using Migo on a remote computer, so Read/Write speeds are critical to productivity. With a slow drive, you end up always waiting for the drive to catch up with your data input (sometimes as long as 45 seconds to a minute) and productivity crawls! As I mentioned, I also own an 8GB Voyager (non-GT) and although the benchmark testing between the drives shows they perform pretty similarly, in actual fact the 16GB GT is much slower! :(: Should we all expect that these larger drives will take a huge performance hit? To eliminate the possibility that the platform I'm using has influenced the results (I'm normally using a box that has a Sabrent PCI add-in card that supplies the USB 2.0 ports), I will try over the next couple of days to employ another box that has USB 2.0 ports run off the MB. I'll keep you posted. :confused: Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted February 14, 2008 Corsair Employees Share Posted February 14, 2008 Are you sure that you have a 16 GT FV, I have not seen them yet? Is you drive Blue and Black or Red and Black? Link to comment Share on other sites More sharing options...
Doogie Posted February 14, 2008 Author Share Posted February 14, 2008 Yes, it IS the GT...black and red, says "Flash Voyager GT" on the drive and on the package. Windows also recognizes it as a "Corsair Voyager GT" :roll: As I mentioned before, this is just a "heads up", as I'm not yet convinced that the problem lies with the drive. I've been using a Sabrent USB 2.0 PCI add-in card to supply my USB ports...and as you are probably aware, not all USB 2.0 ports are the same. Yes, they "should" all conform to the USB 2.0 standards; however, many users don't actually test these add-in cards to see if they perform up to spec unless/until they begin to see a problem. I'm going to use the drive in another box (one whose motherboard supports USB 2.0) for a couple of days to see if there's a difference...I'll keep you posted! :cool: Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted February 14, 2008 Corsair Employees Share Posted February 14, 2008 I checked with our Flash Product manager and he told me that we have not shipped them yet. Can you tell me the reseller you got them from and or provide a link? Link to comment Share on other sites More sharing options...
Doogie Posted February 14, 2008 Author Share Posted February 14, 2008 Sure, it was shipped directly from Corsair in SoCal Link to comment Share on other sites More sharing options...
Doogie Posted February 14, 2008 Author Share Posted February 14, 2008 At home, I've been using a Sabrent USB/PCI card that employs an NEC host controller and at work I use a Toshiba Satellite that employs a Gensys Logic PCI to USB controller...neither of these can produce a decent throughput with the GT (i.e. 17MB/s Read & 7MB/s Write). I recently tried another system that employs an Intel 82801DB/DBM host controller and got better throughput...does Corsair have any recommendations regarding a specific USB/PCI add-in card? :confused: Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted February 15, 2008 Corsair Employees Share Posted February 15, 2008 Doogie Corsair is in the bay area northern California and they have told me we have not sent any out again. Please call and ask for me at 800-205-7657 Link to comment Share on other sites More sharing options...
Doogie Posted February 15, 2008 Author Share Posted February 15, 2008 Yes, it's a genuine 16GB Voyager GT...here's an image of the package... http://img209.imageshack.us/img209/2870/img0002ct3.th.jpg :eek: Link to comment Share on other sites More sharing options...
milnalla Posted February 15, 2008 Share Posted February 15, 2008 While, of course, I could be mistaken, this blister pack does remind me of the market stall knock-offs back home in Manila. Unless Corsair's design department decided to go with less type and more cheesiness. :biggrin: Like I said, I could be mistaken. Please let us know how it did in that other box. Link to comment Share on other sites More sharing options...
Doogie Posted February 15, 2008 Author Share Posted February 15, 2008 The SoCal reference in my previous post was my error the shipping container says that it came from: Corsair 46221 Landing Parkway Fremont, CA 94538-6407 Except for the color, the "GT" and the capacity, the blister pack looks identical to the one that I received from NewEgg with my Voyager 8GB. If you take the time, you'll find this same type of blister pack posted on a number of domestic reseller's websites...:cool: Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted February 15, 2008 Corsair Employees Share Posted February 15, 2008 That looks like the drive, and that is our address. Please call me at 800-205-7657 and I will try to help you. EDIT: Please try the HD-Bench software to test the performance. Link to comment Share on other sites More sharing options...
Doogie Posted February 15, 2008 Author Share Posted February 15, 2008 Ram Guy...I've been using HD tune, HD- Tach and ATTO Benchmark to get my performance information... Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted February 15, 2008 Corsair Employees Share Posted February 15, 2008 Yes I know but please try the version I gave you the link to to compare as that is what we use. And we are testing the drive now with those applications to see if we get different results. Link to comment Share on other sites More sharing options...
kd4wat Posted February 16, 2008 Share Posted February 16, 2008 RAM Guy, it's the age old story--the ones who need to know are the last to know...:roll: Let's put it this way--can production release be far away? I'm betting by March 1st we'll see the 16GB GT. BTW, is there any place where full product read/write statistics can be found? Interested in comparing different manufacturers' products. Link to comment Share on other sites More sharing options...
milnalla Posted February 18, 2008 Share Posted February 18, 2008 Well, I guess I was mistaken. :o: I was going off the design of the 8gb voyager GT pack. ;): In any case, I've spent most of February now checking for the release of the 16gb GT. Today, I revised my search criteria to: "corsair gt 16gb" and lo and behold there were a couple of websites selling voyager GT's and survivor GT's in said capacity. Some had reasonable lead times (2 weeks or so) and others had them in stock, ready to ship, et al. Should I believe them at all? To be sure I couldn't find the expected press release from Corsair. And some of the offerings seemed suspiciously cheap: US$100? :eek: I understand that Corsair may not tell me when these 16gb GT's are coming out, but I don't suppose Corsair could tell milnalla if they've truly come out already? Thanks. Link to comment Share on other sites More sharing options...
Doogie Posted February 18, 2008 Author Share Posted February 18, 2008 Yes I know but please try the version I gave you the link to to compare as that is what we use. And we are testing the drive now with those applications to see if we get different results. Well the results are in for the drive bench testing; however, as I mentioned in my original posting it's the real world results (i.e. running Migo off the drive) that most interests me (for those results see the end of this post). Bench test results: Platform 1: Dell Dimension 2400, 1GB RAM, 400MHZ FSB, CPU Intel 2.4GHz Celeron USB Controller: Intel 82801 DB/DSM EHC File System: NTFS HD-Bench (v3.30) (16GB GT) Read: 27MB/s and Write: 13.5MB/s (8GB non-GT) Read: 27.9MB/s and Write: 5.8MB/s HD-Tach (v3.0.4.0)"long bench test" (16BG GT) Read: 27.8MB/s CPU Usage: 19% and RAT: 1.4ms (8GB non-GT) Read: 29.5MB/s CPU Usage: 23.1% and RAT: 2.3ms HD-Tune (v2.55) (16GB GT) Read: 26.2MB/s CPU Usage: 24% and RAT: 1.2ms (8GB non-GT) Read: 27.1MB/s CPU Usage: 23.1% and RAT: 1.3ms ATTO Benchmark (v2.34) (16GB GT) Read: 28.5MB/s and Write: 19.5MB/s (8GB non-GT) Read: 29.5MB/s and Write: 8.8MB/s Platform 2: Toshiba A85-S107, 1.25GB RAM, 400MHZ FSB, CPU Intel 1.4GHz Celeron USB Controller: Standard Enhanced PCI to USB 2.0 Host Controller File System: NTFS HD-Bench (v3.30) (16GB GT) Read: 13.8MB/s and Write: 9.5MB/s (8GB non-GT) Read: 14MB/s and Write: 5.0MB/s HD-Tach (v3.0.4.0)"long bench test" (16BG GT) Read: 14.1MB/s CPU Usage: 14% and RAT: 1.5ms (8GB non-GT) Read: 14.1MB/s CPU Usage: 16% and RAT: 2.4ms HD-Tune (v2.55) (16GB GT) Read: 13.4MB/s CPU Usage: 16.2% and RAT: 1.3ms (8GB non-GT) Read: 13.4MB/s CPU Usage: 15.9% and RAT: 1.3ms ATTO Benchmark (v2.34) (16GB GT) Read: 14.1MB/s and Write: 11.4MB/s (8GB non-GT) Read: 14.2MB/s and Write: 8.4MB/s Platform 3: ASUS A7N266, 1GB RAM, 266MHZ FSB, CPU AMD Athlon XP 2400+ (2.0GHz) USB Controller: ALi Enhanced PCI to USB 2.0 Host Controller (via a PCI "add-in" card) File System: NTFS HD-Bench (v3.30) (16GB GT) Read: 14.4MB/s and Write: 5.3MB/s (8GB non-GT) Read: 14.6MB/s and Write: 4.3MB/s HD-Tach (v3.0.4.0)"long bench test" (16BG GT) Read: 15MB/s CPU Usage: 13% and RAT: 1.2ms (8GB non-GT) Read: 15.1MB/s CPU Usage: 8% and RAT: 1.4ms HD-Tune (v2.55) (16GB GT) Read: 14.1MB/s CPU Usage 11.8% and RAT: 1.3ms (8GB non-GT) Read: 14.2MB/s CPU Usage: 11.4% and RAT: 1.3ms ATTO Benchmark (v2.34) (16GB GT) Read: 15MB/s and Write: 7.3MB/s (8GB non-GT) Read: 15MB/s and Write: 7.4MB/s A couple of notes...on Platform 3, when empolying anything other than v2.15 (older) drivers from the PCI card manufacturers website, there were delayed write failures following any bench test, if the drive was changed and any test initiated. The older drivers (v2.15) did produce a bit slower read/write values, but were at least stable enough to avoid delayed write failures. A second note, although both drives were formatted in the same way out of the box...HD-Tune consistently showed the 16GB drive as "bootable" and the 8GB drive as not bootable? Both drives were new as of 2008 Finally, unfortunately despite the fact that the GT tests faster on Platform 1, I was unable to experience any real differences between the drives (or the platforms) when using Migo on any of the platforms? The same operations were performed during "testing". Data on the drives consisted of a variety of files up to 2.8GB in size and the operations performed varied from reading/writing email to creating Excel worksheets and creating Word documents. In all cases the delays required to allow the drive(s) to catch up with my data input were frequent and uncomfortably long (i.e. 30 seconds to 1 minute)...I could not distinguish a difference between the drives and although there may have been a difference between platforms, I didn't detect it. :sigh!: All of the platforms that I used might be considered by some as "older" (the newst MB was built in 2005) and this should be considered when interpreting the results (i.e. even the bench tests do not detect a real difference when the platform employs and add-in card for USB 2.0 functionality). For older systems, there appears to be no real assurance that a user can depend upon the GT drive performing any better than a non-GT, even though the GT may be capable of better performance. Link to comment Share on other sites More sharing options...
Doogie Posted February 19, 2008 Author Share Posted February 19, 2008 Ram Guy...since you suggested using the HD-Bench tests to parallel what Corsair is doing to qualify drives, I thought that I would provide you with some additional details... 1. Note that on my original post I said that in my initial testing (which did not include HD-Bench) nothing tested above 17MB/s sequential Read. If you compare that to the results I posted yesterday, you'll see that there were later results, on platform 1, that were considerably higher. HD-Bench shows a lot of variation (+/- 20%) from test to test on the same drive & platform. 2. Using only HD-Bench, I performed a series of tests (10 in series) on the 16GB GT drive on Platforms 1 & 2 (both platforms have MB support for USB 2.0), the averaged results were... Platform 1 (Dell Dimension 2400): Read: 23.6MB/s +/- 20% and Write: 9.5MB/s +/- 30% Platform 2 (Toshiba A85-S107): Read: 11.7MB/s +/- 15% and Write: 7.5MB/s +/- 20% 3. Throughout all of my testing, the most consistent results were demonstrated by ATTO Benchmark. Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted February 20, 2008 Corsair Employees Share Posted February 20, 2008 Are you formatting the drive with NTFS? If so can use format them with FAT32? And normally if the results are not consistent that is due to the system not the drive. And I have seen many OEM systems limit the Voltage bellow spec for USB ports which may cause that anomaly. So on the desk Top system I would suggest using the post on the back of the system next to the Printer Port. And or using an externally powered USB Hub may help as well. Link to comment Share on other sites More sharing options...
Doogie Posted February 20, 2008 Author Share Posted February 20, 2008 Are you formatting the drive with NTFS? If so can use format them with FAT32? And normally if the results are not consistent that is due to the system not the drive. And I have seen many OEM systems limit the Voltage bellow spec for USB ports which may cause that anomaly. So on the desk Top system I would suggest using the post on the back of the system next to the Printer Port. And or using an externally powered USB Hub may help as well. Regarding the file format, yes I will reformat both drives and repeat the testing; however, you'll note that on all 3 platforms using all testing except ATTO Benchmark, the Write values are so far below spec (i.e. less than 1/2 what they "should" be) that I don't think this will make any meaningful difference. I would also point out that when you're selling fast large drives (16GB GT) users are more likely to opt for NTFS so that they can handle > 4GB video files...thus, if you really believe that FAT makes a sizeable difference, perhaps you should post specs in both FAT32 and NTFS? Regarding the variation in the testing...The GT shows greater variation than the non-GT on all platforms. The USB ports that I've tested include: Platform 1 (2 front side USB 2.0 Ports) managed off the MB Platform 2 a laptop (3 built-in USB 2.0 ports) Platform 3 (4 USB 2.0 ports on a slide-in 3.5" front port assembly and 3 USB 2.0 ports coming off the back of the add-in PCI to USB card). They all show significant variation from, especially with HD-Bench; however, it seems to be proportional and platform 3 with the lowest Read/Write values shows the least variation. Another thought...the only testing that even comes close to your specs is on Platform 1 using ATTO Benchmark. Overall ATTO shows the most consistent results when repeating tests (i.e. typically less than 5% variation in Write values and almost no variation on Read values)...and HD-Bench shows the most variation. Seems unlikely to me that the problem of variation is a function of platform or port? :confused: Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted February 20, 2008 Corsair Employees Share Posted February 20, 2008 All testing is always done in Fat16 or Fat32 and never NTFS, using NTFS on a flash drive will slow the performance and decrease the life of the drive. The other thing that I see from your testing is the results seem to be consistent with the respective system suggesting that the slower performance may be system related. Do you have a newer Desktop system that is non OEM like a MB from ASUS, ABIT, DFI, MSI or GigaByte that is AMD64 or Intel LGA775 to compare? My point being that in one of your examples the performance is about on par: Platform 1: Dell Dimension 2400, 1GB RAM, 400MHZ FSB, CPU Intel 2.4GHz Celeron USB Controller: Intel 82801 DB/DSM EHC File System: NTFS HD-Bench (v3.30) (16GB GT) Read: 27MB/s and Write: 13.5MB/s (8GB non-GT) Read: 27.9MB/s and Write: 5.8MB/s HD-Tach (v3.0.4.0)"long bench test" (16BG GT) Read: 27.8MB/s CPU Usage: 19% and RAT: 1.4ms (8GB non-GT) Read: 29.5MB/s CPU Usage: 23.1% and RAT: 2.3ms HD-Tune (v2.55) (16GB GT) Read: 26.2MB/s CPU Usage: 24% and RAT: 1.2ms (8GB non-GT) Read: 27.1MB/s CPU Usage: 23.1% and RAT: 1.3ms ATTO Benchmark (v2.34) (16GB GT) Read: 28.5MB/s and Write: 19.5MB/s (8GB non-GT) Read: 29.5MB/s and Write: 8.8MB/s And I have seen a wider variance on non Intel platform systems so your results do not surprise me. We are still doing some testing so I don't have any further to add. And you have to keep in mind if you are comparing an older 8 Gig FV Non GT that was made with SLC technology it will have the best performance. But that is not available any more. Current Non GT 8 Gig FV Voyager will read about 9.1 Mbps and write 3.3Mbps. Link to comment Share on other sites More sharing options...
Doogie Posted February 21, 2008 Author Share Posted February 21, 2008 All testing is always done in Fat16 or Fat32 and never NTFS, using NTFS on a flash drive will slow the performance and decrease the life of the drive. Understood, however you'll find an increasing desire to employ NTFS in these larger drives and if there's a measurable difference in performance with NTFS you should make that clear :roll: The other thing that I see from your testing is the results seem to be consistent with the respective system suggesting that the slower performance may be system related. Do you have a newer Desktop system that is non OEM like a MB from ASUS, ABIT, DFI, MSI or GigaByte that is AMD64 or Intel LGA775 to compare? No argument here either; however, I think it would be more accurate to say that the performance is probably more specifically related to the USB controller chip. Platform 3 with the poorest performance employs an ALi controller that is known to have "issues" with data throughput. The best results are from the Intel controller. Also, apparently port location (you suggested I try different port locations and I did) seems to have no effect. My point being that in one of your examples the performance is about on par: Actually the ONLY on par performance is Platform 1, but ONLY when using ATTO Benchmark...the Write results with HD-Bench SUCK...and that's what you're using to qualify drives! :eek: And I have seen a wider variance on non Intel platform systems so your results do not surprise me. We are still doing some testing so I don't have any further to add. And you have to keep in mind if you are comparing an older 8 Gig FV Non GT that was made with SLC technology it will have the best performance. But that is not available any more. Current Non GT 8 Gig FV Voyager will read about 9.1 Mbps and write 3.3Mbps. Nope sorry, you're wrong about my 8GB Voyager...it's new (bought from NewEgg in Jan 2008)! I'd bet my eye teeth that it has the MLC controller and you should know that from the test results...your older non-GT Voyagers would consistently test Write speeds above 15MB/s... ;): Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted February 21, 2008 Corsair Employees Share Posted February 21, 2008 It may have been old stock but from the results its definitely SLC based. You have made some good points and I will make sure the respective people see them, however the issue presented has still not changed and comes back to different systems may perform differently that has not been a secret. Bottom line the 16GT on one system is performing at spec maybe a bit low by a few MBps but still what I would consider with in spec. However, I have no problem replacing it if you wish but I doubt a new drive will perform any better. The 8 Gig FV you have is above spec and I would keep it as you may have just got lucky with its performance. Let's get it replaced, please use the On Line RMA Request Form and we will be happy to replace it. Link to comment Share on other sites More sharing options...
Doogie Posted February 22, 2008 Author Share Posted February 22, 2008 It may have been old stock but from the results its definitely SLC based. Can't really see why you'd say that? First, your qualifying test (HD-Bench) on the best performing platform shows only a 5.8MB/s Write speed...that's no where near the posted write speeds for the early non-GTs? Second, "old stock" from newegg??? Not likely, they periodically sell out of your drives and this one would have to have been sitting on the self for 9 months after you'd introduced the MLC version! ;): You have made some good points and I will make sure the respective people see them, however the issue presented has still not changed and comes back to different systems may perform differently that has not been a secret. Bottom line the 16GT on one system is performing at spec maybe a bit low by a few MBps but still what I would consider with in spec. However, I have no problem replacing it if you wish but I doubt a new drive will perform any better. The 8 Gig FV you have is above spec and I would keep it as you may have just got lucky with its performance. Let's get it replaced, please use the On Line RMA Request Form and we will be happy to replace it. Hold on there partner, at this point I'm still trying to find our what's really going on and what to reasonably expect from Corsair's drive...I'm not ready for a "trade-in" quite yet...unless you can guarantee me a fully functional SLC based drive that can consistently perform at the 34MB/s Read & 24MB/s Write speeds that your drives posted when they were first introduced into the market! ;): Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted February 22, 2008 Corsair Employees Share Posted February 22, 2008 Again SLC is not an option, there are no SLC drives bigger than 8 Gig in density. If you want to get it replaced then by all means other wise I am not sure what more I can add to help you. Link to comment Share on other sites More sharing options...
Doogie Posted February 22, 2008 Author Share Posted February 22, 2008 Again SLC is not an option, there are no SLC drives bigger than 8 Gig in density. If you want to get it replaced then by all means other wise I am not sure what more I can add to help you. Ram Guy...to avoid confusion. You wrote: "And you have to keep in mind if you are comparing an older 8 Gig FV Non GT that was made with SLC technology it will have the best performance. But that is not available any more. Current Non GT 8 Gig FV Voyager will read about 9.1 Mbps and write 3.3Mbps." My reply was: "Nope sorry, you're wrong about my 8GB Voyager...it's new (bought from NewEgg in Jan 2008)! I'd bet my eye teeth that it has the MLC controller and you should know that from the test results...your older non-GT Voyagers would consistently test Write speeds above 15MB/s..." We're still discussing the 8GB non-GT and You replied: "It may have been old stock but from the results its definitely SLC based." I replied: "Can't really see why you'd say that? First, your qualifying test (HD-Bench) on the best performing platform shows only a 5.8MB/s Write speed...that's no where near the posted write speeds for the early non-GTs? Second, "old stock" from newegg??? Not likely, they periodically sell out of your drives and this one would have to have been sitting on the self for 9 months after you'd introduced the MLC version!" So what I'm saying is that I have a new Voyager 8GB (non-GT) from a high volume reseller (unquestionably MLC based) and no matter what I throw at it, the Write speed is never less than 4MB/s and it most often tests above 5MB/s. That is certainly better than the 3.3MB/S you quote above, or the 2.2MB/s you quote here http://www.houseofhelp.com/forums/showthread.php?t=67027 but no where near the performance level that you post here http://www.houseofhelp.com/forums/showthread.php?t=33550 for the older SLC drives, so what exactly makes you believe that this unit is SLC based??? Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.