cbjwthwm

Member
  • Content Count

    22
  • Joined

  • Last visited

Everything posted by cbjwthwm

  1. cbjwthwm

    OCZ Intrepid 3800 or 37000

    I've used the 3800 in a custom built server as an Intel SRT cache drive (100GB model) on a RAID and it's been trouble free with current firmware for a couple years now. Its prior drive was an OCZ Synapse which had updated firmware and had no reliability issues either over the course of 3-4 years. I also used a 3700 in a CAD workstation with current firmware and it's been reliable too...it replaced a lower tier Intel SSD whose firmware had defect management bugs in that scenario.
  2. It would be nice to see a list of the drive firmware versions involved in your testing, so that it could be determined if drives with weaker showings (eg: the Intel S3520) were running on early firmware releases that may have seen significant optimization since the review tests. The Micron 5100 Eco must have been tested on an early firmware, as this product was very new at the time.
  3. cbjwthwm

    Micron 5100 MAX SSD Review Discussion

    Was this review done using the firmware version seen on the 5100 Max’s label? It’d be nice if that was noted in reviews, and if it’s an early version of a new type of product like this it would be great to see updated test results once the firmware matures. Another idea would be to delay the review until its next firmware update was released and comment on its development, to give more perspective as to whether its performance is continuing to be optimized or just maintained & stabilized etc.
  4. cbjwthwm

    Crucial MX300 SSD Review Discussion

    "Once the cover is removed, one can easily see that the drive uses a Micron controller, Micron DRAM, and Micron NAND packs" That's a Marvell controller, not a Micron one.
  5. cbjwthwm

    Crucial BX200 SSD Review Discussion

    The OCZ Trion 100 is incorrectly referred to though the entirety of this review as the "Triton" 100. Was it tested with its updated 11.2 firmware?
  6. cbjwthwm

    OCZ Trion 100 Series SSD Review

    Awesome thanks, I'm most interested to see if it improves the smaller models as they have less SLC cache and performed substantially worse than the 960GB drive. If you could provide me with your IOMeter profile used for that test, I could also test the drive that I have as well and report if I see differences.
  7. cbjwthwm

    OCZ Trion 100 Series SSD Review

    I'm primarily wondering if the firmware update improved its peak / max ms scores under IOMeter 4k write latency tests. Considering these don't occur on the S10-based drives with MLC, this appears to be a symptom of the controller's handling of TLC refreshes and garbage collection, or possibly flushing of its SLC cache. When testing drives like these (with SLC caches), it would also be helpful to see a 250GB Crucial MX200 (or Micron M600) included for comparison to a higher class MLC-based drive which also includes SLC cache flushing transitions in its operation. Another useful comparison would be the SMI-based ADATA SP550 which is a SLC-TLC hybrid.
  8. cbjwthwm

    OCZ Trion 100 Series SSD Review

    Despite the slow general showing for this product, my hope is that OCZ will bring some firmware improvements to this controller architecture which is quite obviously a Phison S10 from its firmware version format. I noticed in this review Phison finally fixed their lack of QD scaling the database and server testing phases (vs the Corsair and Patriot S10-based products), so changes appear to be underway. Any plans to update the results with the new 11.2 firmware which was released recently? It appears to be mostly related to sleep mode bugs, but their description of "unresposiveness under heavy I/O" may be improvements in general latency which could make push this controller into a more competitive realm vs its main competition (SMI-based drives) at the entry level. Even a general statement here as to whether you saw any significant differences in non-published testing would be greatly appreciated.
  9. Which firmware revision of DSM 4.3 did you test with? The 8k 70/30 max latency graph also lists the 7K4000's as Seagate ES.3's.
  10. cbjwthwm

    OCZ Vertex 3.20 SSD Review Discussion

    What firmware version was on this drive, and what version was running on your regular Vertex 3?
  11. cbjwthwm

    Plextor PX-M5S SSD Review Discussion

    New firmware 1.02 was just released (the first public update despite changelog notes re: 1.01) for this drive. It will be interesting to see if this firmware fixes the major latency spikes seen from these drives during your enterprise style testing, which I have noticed for the last few months when this drive has been included in comparison data. How soon will you be evaluating this drive in your enterprise testbed with the new firmware?
  12. cbjwthwm

    Corsair Neutron GTX SSD Review Discussion

    It'll be interesting to see if Plextor addresses these potential severe slowdown issues with the M5S in future firmware releases, as they traditionally have a strong advantage versus other consumer drives in enterprise loads. Obviously this entry level sync flash M5 SSD isn't intended to be used in enterprise applications, but there are apparently some holes in its firmware design which isn't surprising considering it's currently at initial release 1.00 firmware. I wonder if drives such as the M3S and M3P have similar tendencies, as their enterprise performance has been shown to be impressive under previous enterprise style testing. Are these enterprise tests more extreme than what the M3P / M2P drives were subjected to in the past? You can definitely see the pedigree of the Neutron's firmware design, coming from their experience with the Seagate Pulstar SSDs.
  13. Any update as to whether the new 1.14 firmware has a more optimal performance balance with the 512GB drive?
  14. cbjwthwm

    OCZ Petrol SSD Review Discussion

    I have two of these drives, so I re-ran the update with the 3.12 firmware's toolbox, and it destroyed the first one (black chassis). It completed without error, asked to shutdown and the drive is now bricked. Since I obviously didn't trust that updater anymore, I used the updated toolbox utility distributed with the Vertex 4's new release version of firmware 1.4 to update the second one, which has a silver chassis. The silver drive completed successfully and requested shutdown the same as the first drive, but still worked after rebooting although it just reflashed the drive with the same 3.12 that was on it. I would suggest that you document in your reviews the firmware revisions tested. Both of these drives are "new" RMA swaps for my original two that had a constantly growing number of defects and read errors, which I flashed to 3.12 and shelved until a new update was released since both showed the same small number of reallocated sectors immediately after updating to 3.12 (the stock firmware did not even have a SMART table!). I'd be quite wary of OCZ, they are notorious for releasing imploding drives in these Indilinx product lines (eg: early Vertex Plus's, which was solved after 3.50 but took them ~6 months to release) and manipulating public perception of what you're buying (eg: Indilinx-infused meaning a completely different architecture Marvell controller in the Petrol/Octane/Vertex 4 with custom "Indilinx infused" firmware). It appears to me that they've waited until they have an internal "stable" release to provide you with a drive to review, but this does not reflect the drives everyone has been buying for nearly 6 months and which cannot be "fixed" with any publicly available firmware. If you could document in your reviews the firmware revision they have provided you with, we can tell when they're essentially manipulating the media to gloss over their past mistakes--which is exactly how I'd describe what they did here. Otherwise, it compromises the integrity of what you're doing here as user experiences with these drives are vastly different than what you have seen in your review evaluation. We know from your Vertex 4 1.4RC firmware release evaluation that your reviews are accurate and not knowingly glossing over deficiencies & stability issues, but I think OCZ pulled a fast one in this case. I'm a huge fan of your site since the early 2000's and it's been my #1 storage resource since the days of Eugene...keep up the good work!
  15. cbjwthwm

    OCZ Petrol SSD Review Discussion

    3.12, I intentionally never purchased any of the drives until the new firmware was available.
  16. cbjwthwm

    OCZ Petrol SSD Review Discussion

    Did you have any stability issues with these drives during testing? I picked up two, and had to exchange both of them due to reallocated sectors and eventual read errors. The stock firmware had no SMART table values, so I didn't work with that firmware other than on the first drive prior to updating to see how much of an improvement it produced. As such, I had both drive failures using the updated firmware--but nothing has been released since. I've basically shelved the drives until new firmware is released, but it's been months now.
  17. All the synthetic scores in Iometer are interesting, but it doesn't seem to be translating into performance in the Storagemark access profiles. LSI's optimizing of SF 5xx generation firmwares to eliminate their max latency deficiencies while maintaining strong performance across the board is far more impressive to me. Thus far, OCZ's development looks like a tuner working over a car for dyno results without providing substantial driveability improvements under real world conditions. OCZ appears to be tuning the drive for marketing specs which are modelled after enterprise load patterns, but this is not an enterprise product. To top it off, it appears the drive isn't stable producing these synthetic numbers anyway. OCZ seems more focused on tweaking their latest and greatest for marketability vs stability and practical performance for its intended market. Meanwhile, Petrols with more of their "Indilinx infused" custom firmware on known stable Marvell chipsets (from the M4) are dropping like flies and they haven't updated that firmware in 3+ months. Interesting, but companies producing LSI/SF drives with current firmware support (Mushkin, Patriot, newer Kingston, lines), Intel, and Crucial will be the ones getting my money while OCZ treats data storage products like gamers overclocking their rigs to point of failure.
  18. Where did you get ahold of this, as Seagate still hasn't officially released this firmware on their firmware download page for the drive?
  19. cbjwthwm

    Lsi Megaraid 300-8X woes

    The ceiling of the burst transfer rate is your problem; if you don't have the bandwidth to allow the two drives to simultaneously scale to their full potential transfer rate, your performance will be artifically limited. The issue is likely firmware on the either the controller (more likely) or the drives (less likely).
  20. cbjwthwm

    Maxline lll vs. T7K250

    Yes, Hitachi is one of the most adjustable drives around in that respect, and the utility used for this is the Hitachi Feature Tool utility located here: http://www.hitachigst.com/hdd/support/download.htm Unfortunately, they do not offer an advance RMA system and that is a significant drawback in my books. Western Digital also has advance RMA service. Samsung does not in my experience. I am from Canada as well.
  21. I know there is one available in the clearance section of www.ncix.com.
  22. I have an Adaptec 2940UW which failed a flash update to the latest firmware. Does anyone know how to do a flash recovery on a card like this? I have experience recovering failed flashes on a variety of devices before including video cards, motherboards, and CD drives. I'm looking for some utility or method to force a flash with no existing functioning BIOS on the card. Adaptec wasn't too useful in helping out.