Low 4K scores for Taichi & 960 EVO
Printed From: ASRock.com
Category: Technical Support
Forum Name: AMD Motherboards
Forum Description: Question about ASRock AMD motherboards
URL: https://forum.asrock.com/forum_posts.asp?TID=4981
Printed Date: 12 Mar 2025 at 7:09pm
Software Version: Web Wiz Forums 12.04 - http://www.webwizforums.com
Topic: Low 4K scores for Taichi & 960 EVO
Posted By: GenesisDoes
Subject: Low 4K scores for Taichi & 960 EVO
Date Posted: 28 Apr 2017 at 12:07am
|
Hey, I noticed that some people are getting low 4K scores for their 960 EVO when using the X370 Taichi. Here is my 960 Evo crystalmark score in Ultra M.2 slot:  Now, here are other people's scores. Gigabyte X370 board with 960 EVO:   250GB 960 EVO:  As you can see my 4K Q32T4 scores are quite low in comparison and other Taichi owners at overclocker forum also are having low scores with the 960EVO. I tried every Samsung driver and default Microsoft windows driver and it doesn't change the score higher. Maybe someone at ASRock can look into this? ------------- Taichi X370, Ryzen 1700 @ 4ghz, 32GB G.Skill TridentZ Ram, Samsung 960 EVO nvme, Intel 600P nvme, 2TB Seagate HDD, 2TB Hitachi HDD, 1TB Mushkin SSD, GTX 1080 |
Replies:
Posted By: berner
Date Posted: 28 Apr 2017 at 1:28am
|
I do have the same Problem on my gaming k4 x370 
|
Posted By: GenesisDoes
Date Posted: 28 Apr 2017 at 2:21am
Hmm, so it appears something with ASRock AMD motherboards and the 960 EVO not reaching full potential of speeds. ------------- Taichi X370, Ryzen 1700 @ 4ghz, 32GB G.Skill TridentZ Ram, Samsung 960 EVO nvme, Intel 600P nvme, 2TB Seagate HDD, 2TB Hitachi HDD, 1TB Mushkin SSD, GTX 1080 |
berner wrote:Posted By: datonyb
Date Posted: 28 Apr 2017 at 3:26am
|
here is my scores with a taichi x370 bios version 2 and using samsung drivers evo 850 500gb sata left evo 960 256gb nvme on right http://smg.photobucket.com/user/datonyb/media/diskmark_zpsrmxclai7.png.html" rel="nofollow">  ------------- [url=https://valid.x86.fr/jpg250]  [/url]3800X, powercolor reddevil vega64, gskill tridentz3866, taichix370, evga750watt gold |
Posted By: datonyb
Date Posted: 28 Apr 2017 at 3:35am
|
^while im very impressed with my evo 850 results on the sata connectors i must say im a bit miffed that this nvme drive baerly outperforms it on 4k ^  ------------- [url=https://valid.x86.fr/jpg250] [/url]3800X, powercolor reddevil vega64, gskill tridentz3866, taichix370, evga750watt gold |
Posted By: wardog
Date Posted: 28 Apr 2017 at 4:08am
|
If, if, IF, this an issue, it would be something AMD would need to correct for. I say that to those of you here who know my style, I don't hold any ones water. Not ASRocks(ok, a little maybe  ) nor any ones. In my book, wrong is wrong.Without knowing specific sata drivers, chipset "drivers', and processor, any "random' pic could be misread/misinterpreted. SATA speeds can be affected by soooo many things it is sometime hard to compare. |
Posted By: GenesisDoes
Date Posted: 28 Apr 2017 at 4:26am
We're all using the same motherboard and various Ryzen processors (there's also 3-4 people on overclock forums in the Taichi X370 thread with the same issue). I used normal windows sata / chipset drivers and then did a benchmark. I installed AMD Ryzen chipset drivers and Samsung nvme drivers and did a benchmark. No matter what, it's a low 4K score. ------------- Taichi X370, Ryzen 1700 @ 4ghz, 32GB G.Skill TridentZ Ram, Samsung 960 EVO nvme, Intel 600P nvme, 2TB Seagate HDD, 2TB Hitachi HDD, 1TB Mushkin SSD, GTX 1080 |
Posted By: berner
Date Posted: 28 Apr 2017 at 4:34am
| also there doesn't seem to be a problem in http://www.tweaktown.com/articles/8073/amd-ryzen-ssd-storage-performance-preview/index3.html" rel="nofollow - this test done on a gigabyte board |
Posted By: wardog
Date Posted: 28 Apr 2017 at 11:23am
That's a really sloppy test there on TweakTown. Mind you, ok, this "review' of theirs was published 03/06/2017 Did you see the CDM edition Ryzen was tested, 5.0.2, dates back to 07/27/2015? The CDM that Intel test was run on is 5.1.2, dated 02/02/2016. http://crystalmark.info/software/CrystalDiskMark/manual-en/History.html Never mind the test dates shown on the PCMark tests. Again, I'm not supporting either side. Show me a heads up test. |
Posted By: parsec
Date Posted: 28 Apr 2017 at 1:46pm
Here's my 500GB 960 EVO in my ASRock X370 Killer SLI/AC board: Looking at a benchmark result does not begin to tell the whole story of how it was achieved. Yes, storage drive bench marking has its tricks and tweaks to get better results. We can get some clues regarding what some of the situation is by looking closely at each of the screenshots in this thread. First, if the drive being tested is the OS/Windows C: drive, the results will be lower than a non-OS drive. The benchmark program does not own the PC when it's running, so any IO done by Windows will interrupt the benchmark. Any non-OS drive will not have the benchmark interrupted by other IO. The second and third Crystal screenshots in the first post that are supposed to be from a Gigabyte X370 board, show the drive being tested were the D: and H: drives. Both not the C: OS drive, so the results will be better. The first screenshot has the drive information cut off, so no idea. All the other Crystal screenshots from ASRock boards in this thread are of the C:/OS drive. So not an identical comparison. Second, the 250GB EVO shown (third Gigabyte screenshot in the first post) which is overall exceptional (with a 50MB/s 4K read) shows 0/233GiB used. That is an empty, new SSD, which will always have great performance compared to a SSD that has been used. Empty is more important than new, but new also helps provide a better result. Next the things we cannot see: CPU OC, Memory OC, CPU power saving option settings, Windows Power Plan option settings, and Windows write cache buffer flushing setting. Do we know about any of these things in the superior benchmark results? No. How did I get the 60MB/s 4K Read speed? I disabled Cool n Quiet and C State support in the UEFI. Even though IO is not CPU intensive, the latency caused by the processor cores switching in and out of the high (C6) C State can be seen in benchmark results. I know it shows up with Intel systems and now I know the same is true for AMD Ryzen. Not a surprise. Gigabyte is (in)famous for setting their default UEFI/BIOS options to provide better performance when their mother boards are tested in reviews, with all default settings. Meaning CPU power saving options on Auto is really Disabled. My ASRock X370 board's UEFI and Cool and Quiet, C6 and Global C6 Support Enabled by default. I'm using the AMD Ryzen Balanced Power Plan, but this applies to any power plan. PCI Express Link State Power Management is set to Off. The Ultra M.2 slots use the PCIe lanes from the CPU, so any power saving on those will cause latency and slower speeds. I haven't tried the High Performance power plan, which I know gives the best benchmark results with Intel systems. I will say in general the 4K performance of any SSD seems to be lower with my Ryzen system than my Intel systems. Of course, Intel has had PCIe 3.0 M.2 support since their Z97 boards released in 2014. We are using the first AMD Ryzen NVMe Option ROMs in our board's UEFIs, which should be improved down the road. Here's one fully tricked benchmark of a 960 EVO on an Intel Z270 board:  ------------- http://valid.x86.fr/48rujh" rel="nofollow"> 
|
Posted By: GenesisDoes
Date Posted: 29 Apr 2017 at 2:30am
I booted from a different drive and disabled power saving options, used Windows High performance power profile and also Ryzen Balanced profile but still get same slow 4K speeds.
------------- Taichi X370, Ryzen 1700 @ 4ghz, 32GB G.Skill TridentZ Ram, Samsung 960 EVO nvme, Intel 600P nvme, 2TB Seagate HDD, 2TB Hitachi HDD, 1TB Mushkin SSD, GTX 1080 |
Posted By: CHO
Date Posted: 01 Sep 2017 at 9:07pm
|
In case of 4K Q32T1 settings, R/W performance will be under 400MB/s but if you change the settings to 4K Q32"T4", you will be able to get over 1,000MB/s @Ryzen environment. Pls check if your setting is correct or not on CDM. |
Posted By: parsec
Date Posted: 01 Sep 2017 at 10:09pm
And we have a winner!  Why did I not include this in my first post, as I said to look closely at the screen shots! The "T4", or four Threads setting, in Settings, Queues and Threads, is seen in all of the Crystal results in the first post for the other boards. You can set four or more threads in the Sequential Q32T1 test too, as well as a higher Queue depth than 32, the maximum for AHCI, but not for NMVe. But there are differences in the standard 4K, and Q32T4 4K tests that are not simply caused by using more threads. We have seen the SATA SSD 4K speeds are lower on our boards too, I don't know what apparently is causing that. ------------- http://valid.x86.fr/48rujh" rel="nofollow">
|
Posted By: zlobster
Date Posted: 02 Sep 2017 at 9:29pm
|
" rel="nofollow - Also, did anyone check thermals during the tests? We all know that NVMe devices are really prone to thermal throttling. 4K random, being last of the read tests, mean that the drive itself has been warmed up in advance, hence possible throttling. |
Posted By: parsec
Date Posted: 02 Sep 2017 at 10:34pm
That is a possibility, but in my case it wasn't an issue. Samsung does not provide their white paper for the 960 series, but did for the 950 Pro, and I have a copy. The throttling threshold temperature for the Samsung 950 Pro is 74° C. Beyond that performance drops quickly, at 75.5° C it is at 75%, and at 76.5° C it is at 50%. At 79° C, it is at 10MB/s! HWiNFO shows two temperatures for 960 SSDs, and mine was never over ~45° C for the NVMe controller. I doubt that the 960 series is any different than this, or throttles earlier than the 950 Pro. The main complaints here are the 4K Q32T1 results, which is the second test of the set of four. If thermal throttling was on at that point, the following tests should be worse than the results shown in this thread. But I cannot say that there wasn't any thermal throttling happening in some or all of the Crystal results posted by ASRock users. It doesn't seem to be very much, if any at all. One little trick you can do with Crystal, is to run each test individually. Instead of clicking the All button, click on one of the four tests, like the 4K Q32T1 box. Only that test will run, and you can wait for the SSD to cool off a bit, and then run another. If your 960 is throttling during only one of the four tests, it really has no cooling at all, and you should do something about that. So just run the 4K Q32T1 test and see what you get. Try changing the number of threads to four (T4) in Settings. ------------- http://valid.x86.fr/48rujh" rel="nofollow">
|
Posted By: zlobster
Date Posted: 03 Sep 2017 at 7:56pm
|
Here is what HWiNFO64 says about 960 EVO: Warning Temperature Threshold: 77 °C Critical Temperature Threshold: 79 °C Latest official FW for the EVO, standard Win 10 NVMe driver. I got an EKWB M2 rad and decent airflow around my 960. Temps during tests were at 35 deg C. Here are the results:   What am I missing? Writes are better than reads? Since when? |