• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

3090 secret memory PerfCap

Fuzzb3k

Fuzzb3k

Fuzzb3k

Associate
Joined
23 May 2003
Posts
305
Location
Fleet, Hampshire
Howdy folks. I've got a Gigabyte Aorus Master 3090 and I've noticed that it will throttle due to 'Thermal PerfCap' reported in HWiNFO, but the GPU itself is only 55C.

After digging a little further, it seems as though the memory chips on the rear side of the card might not be being cooled adequately. Whenever I overclock the memory it will instantly throttle the card due to thermal perfcap. It has been reported that the memory chips on the rear side can reach temps of 90C (!). It looks like there is no way to monitor the temperature of the memory chips right now, but it looks like Nvidia do have access to the temperature sensors on the chips and this is feeding into the throttling algorithm.

I'm going to try putting a heatsink/fan on the rear of the card to see if that helps, but this could explain why quite a few people have reported subpar performance with their 3090's. The 3080 doesn't have any memory chips on the rear side, so I guess it doesn't suffer from this particular problem.

Has anyone else noticed anything similar?
 
EK has launched rear blocks to water cooling the memory

must definitely a hindrance on some cards - the memory on my 3090 has error correction kick in over +200mhz. I have no idea what the temps are but the rear backplate feels HOT

I think EVGA cards have sensors on the memory modules
 
Yeah I heard that EVGA have some sort of patent on monitoring the memory chips. Not sure how much truth there is to that, but seems strange that other manufacturers haven't implemented monitoring of the memory chips.
 
I added a spare AMD Spire cooler heatsink that I had lying around to the top of the card and it has improved things massively. It no longer throttles due to thermals even if I overclock the memory to 700MHz. At the moment the heatsink is just sitting on top, but I might try adding some thermal paste underneath to see if I can improve the heat transfer.

Here's a thermal image of my current setup:



Even though the heatsink pipes are blue (colder), the heatsink itself feels very hot - I can't keep my finger on it for more than a couple of seconds.
 
Fuzzb3k said:
I added a spare AMD Spire cooler heatsink that I had lying around to the top of the card and it has improved things massively. It no longer throttles due to thermals even if I overclock the memory to 700MHz. At the moment the heatsink is just sitting on top, but I might try adding some thermal paste underneath to see if I can improve the heat transfer.

Here's a thermal image of my current setup:



Even though the heatsink pipes are blue (colder), the heatsink itself feels very hot - I can't keep my finger on it for more than a couple of seconds.
Click to expand...

Yep even with a EK block on the front side, and EK backplate, the area around the GPU on the back plate is 70-80c, or so says my temp gun
 
It's cause backplates aren't really designed to dissipate any real amount of heat, they're more there for the visuals.

this was a concern of mine when the 3090 was launched, how would they cool the rear memory - the answer is they don't, as long as it doesn't crash they're fine with it, but there is very little overclock headroom because the memory runs like 90c at stock
 
GDDR6x does apparently run hotter than other memory. Measured at 105 degrees in a pro review I saw. Some backplates also help with sag and card rigidity.
 
This an issue with the 3090FE too?

Can’t say I’ve noticed any problem OC’ing the VRAM so far.
 
HRL said:
This an issue with the 3090FE too?

Can’t say I’ve noticed any problem OC’ing the VRAM so far.
Click to expand...

Have you checked HWiNFO whilst running a GPU intensive app? You know you've run into this issue if you see 'Performance Limit - Thermal = Yes' regularly and your GPU clock speed keeps dropping down.

This is what mine looks like when it's throttled due to VRAM temperature exceeding the limit.

 
Fuzzb3k said:
Have you checked HWiNFO whilst running a GPU intensive app? You know you've run into this issue if you see 'Performance Limit - Thermal = Yes' regularly and your GPU clock speed keeps dropping down.

This is what mine looks like when it's throttled due to VRAM temperature exceeding the limit.

Click to expand...

Nope and it hits 2160Hz on air.

Guessing not then. :p
 
HRL said:
Nope and it hits 2160Hz on air.

Guessing not then. :p
Click to expand...

I think the 3090FE is okay. Not seen the thermal issue once and for daily use memory is sat at +600 Mhz fine. ECC for mine kicks in around +900 or something but thermally is fine.
 
Radox-0 said:
I think the 3090FE is okay. Not seen the thermal issue once and for daily use memory is sat at +600 Mhz fine. ECC for mine kicks in around +900 or something but thermally is fine.
Click to expand...

Would you mind sharing how I can test the VRAM for ECC errors?
 
Fuzzb3k said:
Would you mind sharing how I can test the VRAM for ECC errors?
Click to expand...

Basically test you GPU in benchmarks etc and keep checking the point at which your performance starts dropping. So keep increasing the memory in say 100 up to around 500, then in 50 mhz increments, to point performance dips.
 
HRL said:
Sounds like an AIB oversight then TBH, which is a bit worrying really.
Click to expand...

I might take off the backplate later and see if there's any thermal pads on the memory chips. But even if there is, there's no active cooling of the back side of the card as it's not joined in any way to the (massive) front cooler so it's really relying on dissipation from the backplate. Seems a bit rubbish considering all the effort Gigabyte put into the front cooler.

Surely it wouldn't have taken much effort to include a heatpipe or two to hook over to the back side?
 
Fuzzb3k said:
Would you mind sharing how I can test the VRAM for ECC errors?
Click to expand...

I haven't found any easy way. Maybe there is a way to generate error log files for VRAM, but if there is I have no idea how to do it.

The only reliable way we have is benchmarks. Set a base line score in 3d mark timespy with stock clocks - and then you can start increasing memory clocks speed, you'll know when ECC kicks in because the benchmark score will drop a lot. In 3d mark timespy I go from 21000 score at +200mhz on the memory to 15000 score if I go over that due to ECC.

Because of how ECC works, I believe it's beneficial for everyone to run timespy at least once with the card stock when they first get the card so I can set a baseline. Then in future if you suspect something is wrong with the card or you are just curious, you can run timespy again and compare against when you first got the card and if the score has dropped significantly then you know something is wrong. It remains to be seen if prolonged high tempreture on the GDD6X memory will cause the memory performance to degrade over time, but if it does degrade you will see ECC kick in and drop your performance so its a dead give away.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom