|
SourKraut posted:Aren't the WD Blues just rebranded Sandisk X400s currently?
|
# ? Jan 14, 2017 20:44 |
|
|
# ? May 11, 2024 12:45 |
|
Anime Schoolgirl posted:Also overclocked, which apparently hosed everything up This is the first I've heard of that. What kind of issues is it causing? Or even better if you have an article on it I could read. Thanks.
|
# ? Jan 14, 2017 21:58 |
|
not the overclocking specifically but there are bugs in the WD blue SSD's custom (WD's) firmware that I can't pull out at the moment, potato salad can probably illuminate you on this
|
# ? Jan 14, 2017 22:07 |
|
Anime Schoolgirl posted:not the overclocking specifically but there are bugs in the WD blue SSD's custom (WD's) firmware that I can't pull out at the moment, potato salad can probably illuminate you on this Sounds interesting. I'll just google, look at anandtech, and tweaktown etc.
|
# ? Jan 14, 2017 22:10 |
|
I'd be fascinated to see these articles as well.
|
# ? Jan 14, 2017 22:23 |
|
Is there any way to know what you're going to get in a laptop as the OEM drive? Just found a mSATA adapter with an SK Hynix drive in a little 2.5" bracket in this Dell.
|
# ? Jan 14, 2017 23:09 |
|
They should provide complete system specifications in the detailed specifications pages for each laptop.
|
# ? Jan 15, 2017 04:05 |
|
Is it really the case that the Intel 600p NVMe drive is actually much slower than the 850 EVO SATA? Jebers.
|
# ? Jan 15, 2017 18:09 |
|
Bob Morales posted:Is there any way to know what you're going to get in a laptop as the OEM drive? Just found a mSATA adapter with an SK Hynix drive in a little 2.5" bracket in this Dell. No, at least not in my limited experience. I got an MS GS73VR and the M.2 slot is set for SATA or x4 NVMe. Mine came with an SKHynix SATA drive but I've seen different reviews of the computer with Samsung and others.
|
# ? Jan 15, 2017 18:11 |
|
Bob Morales posted:Is there any way to know what you're going to get in a laptop as the OEM drive? Just found a mSATA adapter with an SK Hynix drive in a little 2.5" bracket in this Dell.
|
# ? Jan 15, 2017 18:23 |
|
Have a good relationship with your var. No, seriously, you need an inside agent looking into it for you.
|
# ? Jan 15, 2017 18:33 |
|
I've seen a lot of benchmarks showing zero or tiny improments going from a SATA SSD to an NVMe SSD even in things like game loading. Is this a case where the software hasn't caught up to the hardware or does having faster and lower latency storage beyond SATA SSDs really not help games load faster? I've seen cases where a game load took 30+ seconds with both the SATA and NVMe drives, that's a really long time by modern PC standards. What is going on that takes up that much time and yet isn't improved by faster storage?
|
# ? Jan 15, 2017 20:19 |
|
MaxxBot posted:I've seen cases where a game load took 30+ seconds with both the SATA and NVMe drives, that's a really long time by modern PC standards. What is going on that takes up that much time and yet isn't improved by faster storage? I would guess these actions are CPU limited - such as the CPU can only load or unpack so many assets so allowing it to access more doesn't help.
|
# ? Jan 15, 2017 20:46 |
|
this kind of poo poo is why i never buy any laptops but lenovo thinkpads and asus/msiMaxxBot posted:I've seen a lot of benchmarks showing zero or tiny improments going from a SATA SSD to an NVMe SSD even in things like game loading. Is this a case where the software hasn't caught up to the hardware or does having faster and lower latency storage beyond SATA SSDs really not help games load faster? I've seen cases where a game load took 30+ seconds with both the SATA and NVMe drives, that's a really long time by modern PC standards. What is going on that takes up that much time and yet isn't improved by faster storage? hilariously the only game that doesn't gently caress itself over on an SSD is doom 4 and mostly because it loads 6 gigs of assets per level so it still takes like 20 seconds to load, it just takes 2-3 minutes on a hard drive Anime Schoolgirl fucked around with this message at 20:50 on Jan 15, 2017 |
# ? Jan 15, 2017 20:46 |
|
Anime Schoolgirl posted:this kind of poo poo is why i never buy any laptops but lenovo thinkpads and asus/msi What's wrong with this? It's either a full SATA drive or this adapter with an mSATA drive. There's no performance difference.
|
# ? Jan 15, 2017 21:00 |
|
metallicaeg posted:What's wrong with this? It's either a full SATA drive or this adapter with an mSATA drive. There's no performance difference. also the fact that you can't fix most dell laptops nowadays without literally breaking them as I've learned over the last week
|
# ? Jan 15, 2017 21:06 |
|
MaxxBot posted:What is going on that takes up that much time and yet isn't improved by faster storage? Some back of the envelope estimates suggest it's decompression limited. (i.e. CPU) Some numbers: JPEG seems to decompress at maybe 20 MB/s of compressed data per core. gzip at about 80 MB/s, I think. With 8 cores, that's about SATA speeds. It'll be interesting to see what happens when this bottleneck becomes more recognized. Maybe nobody cares? Decompression accelerators? Compression methods with faster decompression? GPU assisted decompression? FPGAs? (Hey, there's reason to want them on servers, and there may be reason to want them on cell phones, why not a reason to want them on the desktop!)
|
# ? Jan 15, 2017 21:29 |
|
crazypenguin posted:It'll be interesting to see what happens when this bottleneck becomes more recognized. Maybe nobody cares? Decompression accelerators? Compression methods with faster decompression? GPU assisted decompression? FPGAs? (Hey, there's reason to want them on servers, and there may be reason to want them on cell phones, why not a reason to want them on the desktop!) Why not just do the unpacking the first time the game is played? Saved the uncompressed assets to the local storage.
|
# ? Jan 15, 2017 21:37 |
|
I think just because disk space still isn't cheap. Modern games suck up 10s of gigabytes compressed. Completely uncompressed would be almost 10x that, no?
|
# ? Jan 15, 2017 21:47 |
|
That wouldn't be the best idea.
|
# ? Jan 15, 2017 21:47 |
|
crazypenguin posted:Some back of the envelope estimates suggest it's decompression limited. (i.e. CPU) That's pretty insane, LZO 2012 was decompressing at 1.2 GB/s on sandy bridge. Why would anyone gimp their performance by using a slow algorithm?
|
# ? Jan 15, 2017 22:57 |
|
Harik posted:That's pretty insane, LZO 2012 was decompressing at 1.2 GB/s on sandy bridge. Why would anyone gimp their performance by using a slow algorithm? A couple things: I think jpg decompresses at closer to 80mb/s these days than 20mb/s That said, a 1mb jpg file is going to compress at say 10:1, so you end up with 10mb of raw pixel data. So you might only be uncompressing 10mb of data per second but you end up creating 100mb per second. Second, generic data compression algorithms don't compress images very well, and they're are just working on a stream of data. JPEG is analyzing blocks of an image and all that poo poo.
|
# ? Jan 15, 2017 23:38 |
|
Bob Morales posted:Is there any way to know what you're going to get in a laptop as the OEM drive? Just found a mSATA adapter with an SK Hynix drive in a little 2.5" bracket in this Dell.
|
# ? Jan 15, 2017 23:57 |
|
Bob Morales posted:Why not just do the unpacking the first time the game is played? Saved the uncompressed assets to the local storage.
|
# ? Jan 16, 2017 00:21 |
|
I think uncompressing textures is the least issue of all. Parsing geometry data and such takes probably more time.
|
# ? Jan 16, 2017 00:56 |
|
Harik posted:That's pretty insane, LZO 2012 was decompressing at 1.2 GB/s on sandy bridge. Why would anyone gimp their performance by using a slow algorithm? I suspect you're thinking of the wrong number. Compression algorithms are usually measured on rate of decompressed output per second. This is generally a good number to optimize and compare algorithms on: it reflects both improved compression ratio and improved decompression speed, both. (That is, if my algorithm is 10X slower, but achieves 100X better ratios, then I win.) I was trying to give numbers for the rate these algorithms can consume input (compressed) data per second. This is generally a useless number as a point of comparison between algorithms, so it's very rarely reported or measured. But it's exactly what we care about when looking at why NVMe isn't a performance win. We can load the data faster than it can be decompressed. I looked up LZO, and it looks like there may be licensing problems. However, it looks like Google has released "snappy" which I can't find good enough numbers on to figure out at what rate it can generally decompress data. But I suspect it was designed specifically with "oh poo poo, SSDs are CPU-bottlenecked on decompression, let's optimize for that while avoiding patents." So maybe if games ditched zlib for snappy, we might cut loading times in half. But (take this with a grain of salt: I can't find numbers) I think we'd still be CPU limited, actually.
|
# ? Jan 16, 2017 01:12 |
|
Bob Morales posted:A couple things: I think jpg decompresses at closer to 80mb/s these days than 20mb/s I meant mesh data, which is generic. Point taken, our figures are probably at opposite ends of the algorithm.
|
# ? Jan 16, 2017 01:58 |
|
I couldn't help but think about this more. I found a nice set of benchmark numbers, which can be sorted by decompression speed: https://quixdb.github.io/squash-benchmark/#results-table It looks like a reasonable sweet spot is the lzo/lzo1b option, which gets about a 2.4 ratio with 560 MB/s decompression output on one core. So, assuming linear scaling (hah), that's able to consume 930 MB/s of input on a 4 core cpu. NVMe drives can firehose 3500 MB/s at it, so yeah. Definitely still very CPU bottlenecked. But SATA is about 600 MB/s, so a good 50% improvement. The same benchmarks put zlib (the compressor that e.g. skyrim/fallout uses) at 200 MB/s at 3 ratio, or consuming 270 MB/s at 4 cores. This game also stores textures as dds files in S3 format, though, which doesn't need additional decompression before being uploaded to GPU. ...Except they might be also recompressing those with zlib, I dunno. Meanwhile, Xilinx advertises 4000MB/s consumption speeds (100 Gbps, zlib's 3 ratio) on cheap fpgas for zlib: https://www.xilinx.com/products/intellectual-property/1-7aisy9.html (actually, it looks like that's specifically talking about compression, but that should be a good conservative proxy for what's possible.) So an FPGA can easily handle full speed NVMe drives, and without having to choose a less efficient compression algorithm, either. So hardware can definitely handle this bottleneck problem. But like I said, we'll see whether this is regarded as a real problem, and who knows how it might get solved, if so.
|
# ? Jan 16, 2017 03:36 |
|
redeyes posted:Is it really the case that the Intel 600p NVMe drive is actually much slower than the 850 EVO SATA? Jebers. No? They kind of trade blows on stats - PCIe gives the Intel an edge on sequential reads as long as you have one of the larger-capacity models and possibly sequential writes too, and the Samsung's controller is capable of more IOPS so it tends to pull ahead on random loads. As far as subjective "feel" goes, I have the 512GB 600p in my laptop and the 500GB 850 EVO in my desktop and I honestly can't tell a difference between them unless I use a benchmark. Note also that the 600p's stats initially came out showing the same durability in TB written across all models. A lot of people pointed out that this was nonsensical and Intel released new numbers that looked proportionally better for the larger models, so if what you're looking at shows an oddly low durability for the 600p make sure you're not looking at the original stats. Eletriarnation fucked around with this message at 06:37 on Jan 16, 2017 |
# ? Jan 16, 2017 06:35 |
|
Same for the 960 Pro vs 850 EVO. The former has like six times the read throughput of the latter (3GB/s vs 0.5GB/s), but in application and game benchmarks, it ain't reflected.
|
# ? Jan 16, 2017 14:05 |
|
Eletriarnation posted:No? Gotcha. Thanks, 600p inbound.
|
# ? Jan 17, 2017 23:40 |
|
the 1500mb/sec burst reads are mighty nice anyhow
|
# ? Jan 18, 2017 00:10 |
|
Anime Schoolgirl posted:the 1500mb/sec burst reads are mighty nice anyhow It's a good price for NVMe for sure. Hopefully it wont bite me in the rear end over some odd write-blocking problem.
|
# ? Jan 18, 2017 02:04 |
|
http://www.tomshardware.com/news/toshiba-wd-semiconductor-nand-spin-off,33448.html
|
# ? Jan 20, 2017 09:24 |
|
What is the current drive of choice for NVMe on a budget?
|
# ? Jan 20, 2017 09:49 |
|
Lovable Luciferian posted:What is the current drive of choice for NVMe on a budget? The Intel 600p and/or the Plextor M8 series. I'd highly recommend going with the heatsinked version of the Plextor. Toshiba and OCZ also make a decent drive. There'll be a ton of new drives in the next six months, though.
|
# ? Jan 20, 2017 09:52 |
|
Anime Schoolgirl posted:http://www.tomshardware.com/news/toshiba-wd-semiconductor-nand-spin-off,33448.html That's interesting stuff! Although Seagate is doing a fair amount of nvme flash stuff but mostly just enterprise. WD is positioned super well and having SanDisk, WD and HGST brands gives quite the range from low end consumer to enterprise. I am biased because my company sells controllers to wd/hgst, now hopefully they don't pull a samsung and start making their own.
|
# ? Jan 20, 2017 11:06 |
|
Tom's Hardware did a 4KB failure/stress test on the 512GB 600p: http://www.tomshardware.com/reviews/intel-ssd-600p-nvme-endurance-testing,4826.html
|
# ? Jan 20, 2017 11:09 |
|
Okay, so I got my new Dell 3000 and was hoping to put an SSD in it, but the internal layout is a bit odd to say the least. There's some PCIE x1 slots free though, would they be any use in adding an ssd to the computer? I was looking at adaptors like these for example: https://www.newegg.com/Product/Product.aspx?Item=N82E16815124167&cm_re=ssd_adaptor-_-15-124-167-_-Product https://www.newegg.com/Product/Product.aspx?Item=9SIA1JM33B7846&cm_re=ssd_adaptor-_-9SIA1JM33B7846-_-Product Or just mount the drive onto the card. Is there anything I should consider before getting one of these?
|
# ? Jan 20, 2017 11:16 |
|
|
# ? May 11, 2024 12:45 |
|
priznat posted:That's interesting stuff! Although Seagate is doing a fair amount of nvme flash stuff but mostly just enterprise.
|
# ? Jan 20, 2017 11:38 |