RTX 2080 ti low read speed
Hai!Please tell me, I recently bought a new vidyuhi and tested it, and in Hades it gave out a very small Memory Read 3130 MB / s, although if you compare it by videos on YouTube, others have this Memory read at the level of 12000 MB / s
Tell me fraternally what is not so😅
Here is my GPU test from aida
PapunLapatun
Be able to install the full version of AIDA64, otherwise you look funny with your "proofs"!
SonyK_2
AIDA64 Engineer Edition 6.60.5900 Portable by FC Portables [Multi/Ru]
Update date: 04/06/2022
Official site: AIDA64
Official website of the collector: fc portables
Treatment: not required
Portable (portable) version from FC Portables.
Does not require installation.
- what is your opinion about this AIDA64?
I also forgot to say: I have 4 computers, they work through a KVM switch. Maybe it affects?
vftor
vftor wrote:
AIDA64 Engineer Edition 6.60.5900...
On the TS screen...
https://forums.playground.ru/hardware/video/malenkaya_skorost_chteniya_po_pamyati_rtx_2080_ti-1013267/#comment-19119777
..., mine...
https ://forums.playground.ru/hardware/video/malenkaya_skorost_chteniya_po_pamyati_rtx_2080_ti-1013267/#comment-19119786
..., and another user...
https://forums.playground.ru/hardware/video/malenkaya_skorost_chteniya_po_pamyati_rtx_2080_ti-1013267/# comment-19119905
...you can see the program version is 6.70.6000. I have a Chupacabra repack, the Extreme version is installed.
Spoiler 
... I have 4 computers, they work through a KVM switch. Maybe it affects?
It seems to be for the keyboard and mouse, what is the connection with the performance of the video card?
SonyK_2
I downloaded AIDA64 - 4 options:
- AIDA64 Extreme ... v.6.50.5800 (virus)
- AIDA64 Extreme Edition 2020 v.6.25.5400 (virus)
- AIDA64 Engineer Edition 6.60.5900 Portable (without installation)
- AIDA64 5.98 .4800 RePack (&Portable) Extreme, Engineer, Business, Network Audit (10/10/2019) - if this version is Extreme, maybe it will give more accurate results?
SonyK_2
SonyK_2 wrote:
It seems to be for the keyboard and mouse, what is the connection with the performance of the video card?
It toggles Monitor, Keyboard, Mouse and Sound. It has a monitor resolution limit of 1920x1080. 60Hz
And one more thing: on a PCIe 3.0/2.0 x16 motherboard, but the video card works on PCIe 2.0 (slightly reduces performance) due to the i7 2600K CPU, and to work on PCIe 3.0, you need another CPU
vftor
vftor wrote:
... downloaded AIDA64 - 4 options ...
Yes, at least 100500 - if we compare, then on the latest version, which is on the screenshots of others.
...virus... virus... I wrote
about this above, read the posts in the topic more carefully : 3.0/2.0 x16, but the video card works on PCIe 2.0... to work on PCIe 3.0, you need another CPU and test the speed of the 960th on it. It's easier to search the web for screenshots with test results for the 980 Ti and similar CPUs.
Lex-one
Yes, in any case, people with the same settings, the indicators are at the level of yours. The card is working at full capacity, everything seems to be in order, but for some reason it gives a result half less than that of you and other 2080 ti owners
MOUSE lapwing
Returned and read your post, with a standard frequency for her, the temperature does not exceed 70 degrees, mostly 68. Thanks for the advice!
The power supply for the corsair is 850w platinum
No matter how everything works, the card does not make noise, does not heat up, the frequency keeps its native, it loads at 100% in terms of power and outputs everything as stated by the manufacturer, BUT for some reason the results are such that it seems to work by 50% 🤓
PapunLapatun
SpoilerUnderstood the software update? Also, pick up a game.
A similar garbage (loss of performance at full load) is observed when installing a video card in a PCI-e x16 slot version X4. This is a stupid note, but anything already happens. Tell me what kind of boards and where it is (they even washed down the photo, we'll see the map later). Also, I would like to see, but this is not a fake, stitched under 2080Ti, bought on the Chinese market from a noname. And finally, does it even correspond to the power consumption?
Check at least the version under load:
Spoiler 
And yet, reinstall to a clean driver, with removal via DDU
PapunLapatun wrote:
less than you and other owners of 2080 ti
He made a mistake:
SonyK_2 wrote:
And nothing that he has anti-aliasing settings on Furmark's screen set to 4X (AA: 4 samples), but on yours it's disabled (AA: 0 samples)?
Therefore, the indicators do not converge.
PapunLapatun wrote:
You have no idea what the problem
was😂😂😂 OLD DRIVERS🤓🤓🤓
Well. It happens. Something is already the third theme of the guys with the conflict of drivers. Order then.
01001000 01101001
You have no idea what the problem
was😂😂😂 OLD DRIVERS🤓🤓🤓
Now I will send all the results in the post below
01001000 01101001
SonyK_2
Lex-one
vftor
As a result, this whole problem was not updated video card drivers
Tests in Aida (still not in the full version, do not scold plz) grew to those of other RTX 2080ti owners
Tests in Bagel, respectively, also “tightened up » to the required
PerfCap in white color when loading (idle)
Here is a screen after updating the system
I want to separately say thank you for your help in lowering the temperature to normal levels
Thank you for solving the problemðŸ‘🤓
As for games, the indicators also pulled up to those that can be seen on YouTube benchmarks.
PapunLapatun
No, your problem was not in the driver, since you had the same driver.
You probably have x4 msaa anti-aliasing in your nvidia control panel.
Lex-one
Lex-one wrote:
...the problem was not in the driver, because you had the same driver.
Maybe not in the driver, but on the first screenshots he has 512.95, and on the last 516.40.
Spoiler
SonyK_2
I also had 512.95, 516 came out yesterday and he checked the clean installation box.
That's all.
Lex-one
If you have the desire and the opportunity, can you run two tests in a Bagel in x4 and x8 smoothing?)
I'll run mine and check it just for the sake of experiment
If possible)
PapunLapatun
Lex-one wrote:
...516 came out yesterday and it ticked clean install.
As for the "clean installation" it is better to check with him.
PapunLapatun
PapunLapatun wrote:
...can you run two tests in a donut in x4 and x8 smoothing?... I'm sorry
to interfere, but why do 2 tests in a "donut"? - just agree on the settings and do 1 test on exactly the same (including resolution). I usually set mine to the maximum (in full screen mode) for 10 minutes (add 0 in the duration settings) ...
Spoiler 
... although a minute is quite enough for a normal comparison (it is by default).
Spoiler