Recent content by DarkServant

@jcizzo I think that you have read the prerequisites for a reliable ZFS storage system? You should not go with a base-system without a good amount of ECC-DRAM. Storage-wise, you have heard about enterprise SSD's which do have some extras like powerloss-protection (data-in-flight! not only at...

I've never read from someone who has done a FW flash to an OEM (vendor-locked) drive with a standard firmware and succeeded... I see a lot of SSD's on ebay which are "vendor-locked" (sixth digit/character not 0 where it should be, only some pm17xx have a B instead in think). Be careful what you...

Never heard of it... But there is an E2 standard coming, which allows 64+ NAND-packages on a single drive, plus enough space for PLP-caps (200 x 76 x 9,5 mm). On u.2 it's full at about half the packages with two PCB's. And in case of thermals it seems not to be a great idea to have 2 PCB's over...

Is u.2 not an option? With u.2 you have more options to mitigate heat-problems IMHO. With the heat-sink for m.2 22110 it's probably clever to get some height measurements of the components (Controller, NAND, DRAM-cache and PMIC -> they can get hot too), to apply the correct thickness of...

I have never heard of an an JXTD***Q hardware revision.The pm893a which probably has the newer 133-layer v6 prime NAND uses JXTG***Q. Stick to the JXTC***Q when this is the designation on your drive. Nearly all the updates are NF (NoFormat) even if the _NF on the file is missing. Or ask the...

Hi, it would be nice to create a repository for Toshiba/Kioxia SSD's. They are too on the list for do not buy because of that, but they have a solid quality hardware. PS: An FL6 with 3,2TB sounds really nice.

Hi there Does anyone know exactly how thick the thermal-pads on the RTX 6000 Ada and probably RTX Pro Max-Q 6000 are? I think it's 1mm for VRAM modules and 2mm for the VRM's (DriverMOSFET's), but on the Backplate i have no idea. On the consumer variants it's more like 1,5mm for the VRAM...

I looked at the file with a hex-editor and it seems that this is the firmware file itself. Always check at least the header and fly over it, to see if it's not garbage. Some infos at the start, then follows the encrypted-sign-part and after this the fw-code follows. It has a lot of free space...

Hi, i asked the vendor about this problem and is was consistent wich another sample, and all the professional graphics cards are assembled by Nvidia themself themselves. But anyway, the answer to the problem was more than unsatisfactory: Sounds like always, when big companies try to cover up...

In GPU-Z it is as it should be according to specs: And you are right, this ECC is definitively not how ECC should be done on a hardware-level, no additional memory chips, probably some kind of cheap Inline-ECC... ECC disabled gives about 120GB/s more VRAM throughput. This GPU's are only about...

I cannot belief this... the last 1280MB seems to be affected by a heavy slowndown in this test... this is/should be a full unlocked working AD102 >:-( Tested under Windows 10 IoT LTSC 21H2 with the latest updates. Nvidia driver version"QNF" 576.02 Is there the VRAM not backed by the L2-Cache or...

There is a SSD-Database at TechPowerUp: SSD Database TechPowerUp® It can contain some valuable informations, as there are not only consumer SSD's listed. And some videos about NAND in general can be found on YouTube, like how it works... pages, blocks, planes, peripheral circuits like...

yeah, i have too some five sv843 (the same internals as the 845dc pro, but less fixed OP, 960GB instead of 800GB) one Samsung and the others are HP branded. They should be working in the NAS which i planned to build, but as the time goes by, they where too small and the compact Supermicro case...

I would recommend updating the SSD's, it does *should* not destroy the data, but as always back up the data if any important is on the drives. You could overprovision the drives to 1,6TB via HPA to gain more write performance and more write endurance (can lower the write-amplification a lot)...

With the capacity/performance, i've seen this over and over in datasheets, there is mostly some sweet spot. But why, i don't know... can be limited channels of the controller, or different NAND sub-generations (Samsung uses slightly older NAND on the highest capacity drives)? On SAS i had only...