• 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.

Inno3d Ichill X3 RTX 4090

Adrian1976

Adrian1976

Adrian1976

Associate
Joined
17 Mar 2017
Posts
947
Location
Manchester
Hi,

So I have had this GPU since release and all has been well however I have always had it at 80% power limit and the hotspot generally reaches a max of 90c.

But after seeing others post about the hotspot on the Ichills has got me a little concerned as to whether I should get this returned to OCUK because after testing it today at stock 100% power settings and even a more aggressive fan curve when this GPU is reaching 445w I am seeing the hotspot peak at at over 101c regularly.

Now as far as I am concerned I know some will say this is normal behaviour but when the core is not going over 78c that's almost a 23c delta and 100% a contact or paste issue as one user on Reddit has repasted his and now he's not seeing more than a 15c delta so you would say poor manufacturing on Inno's part or poor assembly more like.

Now I am not prepared to strip down a £1700 GPU no matter what, As far as I am concerned any product costing this amount of money should be built properly or assembled properly, I am more concerned about the longevity of this GPU if I was running it at stock in a very good well ventilated case and we are currently in the middle of winter in the UK so this is going to get worse in the summer.

Anyone know how long the process if I send this back to OCUK, Is there a faster option or any advanced RMA on Inno's part if dealing with them or is it better to stick with the retailer?

Thanks in advance.
 
Last edited:
Ooops for to add the image.
8YC7wyY.png
 
Last edited:
Adrian1976 said:
Anyone know how long the process if I send this back to OCUK, Is there a faster option or any advanced RMA on Inno's part if dealing with them or is it better to stick with the retailer?
Click to expand...
Might be best to speak to OcUK directly about this, Inno may deem this as 'normal' behaviour and within their parameters (which are sometimes v loose from manufacturers).
 
That's very high delta and hotspot temp. Probably bad contact between GPU die and cooler or dried out thermal paste. Best to contact retailer and ask for RMA, as changing paste yourself might invalidate warranty.
 
Yeah I am sure Inno would say this is normal but we all know that size of delta and these hotspot temps are not normal behaviour and I'm sure OCUK knows this is not normal either, Especially after seeing other 4090s nowhere near this hotspot temp at stock.

I even tried 100% fan speed it made 0 difference so does sound 100% a contact issue to me.

Other Inno users are reporting a similar trend also so my worry is if I go through the RMA process and end up with a replacement doing exactly the same things because it's just how Inno assembles these GPUs poorly.

Not sure what to do right now.
 
I'm in a similar boat with my 6800 XT. The delta's always been quite large (mod to high 20's) but temps were good, but over time the hotspot is now close to 100°c so I think I'm going to repaste. The difference is mine is out of warranty so I'm not voiding anything.

Definitely speak to OCUK but as the temps are within "normal" ranges (not throttling) then I agree that Inno3D wouldn't see it as an issue.
 
Kirby Wurm said:
I'm in a similar boat with my 6800 XT. The delta's always been quite large (mod to high 20's) but temps were good, but over time the hotspot is now close to 100°c so I think I'm going to repaste. The difference is mine is out of warranty so I'm not voiding anything.

Definitely speak to OCUK but as the temps are within "normal" ranges (not throttling) then I agree that Inno3D wouldn't see it as an issue.
Click to expand...
I’ve also found that hotspot temps deltas are wider on AMD Vs Nvidia. I suspect this is due to the differences in architecture design and that Nvidia GPU die is usually much bigger in size.
 
Kirby Wurm said:
Definitely speak to OCUK but as the temps are within "normal" ranges (not throttling) then I agree that Inno3D wouldn't see it as an issue.
Click to expand...
Not sure OCUK would do anything, really. If the manufacturer says it's normal, OCUK will accept that fact.

I guess if he's had the card less than a couple of weeks he might be able to return it, otherwise just have to live with it, no?
 
Yeah I mean it seems to be performing ok, One thing I didn't check was the clock speeds when it hit 101c to see if it was throttling or not but the fact we're in winter currently there's only one way this hotspot temp is going in summer and that's rising further so I'm pretty sure it may throttle if it isn't already.

When it's power limited to 80% the hotspot is around the 85c range which imo is normal for this power of GPU as I'm seeing most other 4090's on air averaging around that point on gaming full load but the fact I'm hitting 85-86c at 80% power limit is not great is it.

Don't know whether to live with it and if I still have it after the warranty repaste myself or see what OCUK say.

Be interesting to see if there's any other Ichill X3 owners on here who are seeing a similar temp or not.
 
I've got the X3 OC. I haven't seen the hotspot hit 101c, but I've seen it in the high-90s in Total War: Warhammer 3. That game can push system temperatures higher than any synthetic benchmark I've seen (I ended up capping it at 60fps because I didn't like that). The delta's only around 15 degrees then, though.

In every other game, I'm getting hotspots in the mid 80s and main temps around 10 lower.
 
Last edited:
Rogueywon said:
I've got the X3 OC. I haven't seen the hotspot hit 101c, but I've seen it in the high-90s in Total War: Warhammer 3. That game can push system temperatures higher than any synthetic benchmark I've seen (I ended up capping it at 60fps because I didn't like that). The delta's only around 15 degrees then, though.

In every other game, I'm getting hotspots in the mid 80s and main temps around 10 lower.
Click to expand...
Interesting to know thanks, Yeah I don't expect a perfect 10c difference between core and hotspot but 23c difference at one point imo is not acceptable.

The fact mem temps are really good on this I suspect either the Vram pads are slightly too thick for these Inno GPUs for the core to contact properly, I've seen this issue with my 3080 where if either the Vram pads are too thick or too rigid the core doesn't make the best contact.

Either that or the GPU screws have not been tightened properly, I'll try giving them a little turn next time when I take it out eventually when my ATX 3.0 PSU finally arrives.

Don't think I'll bother sending it back, I may try and resolve it myself at some point or just live with it, Annoying though I'll be honest, Shouldn't have to mess around with a new GPU because assembly is crap.
 
Last edited:
remove the custom fan curve you have set and let it run default so it is either going to be louder or cook itself
 
Last edited:
Alt0153 said:
remove the custom fan curve you have set and let it run default so it is either going to be louder or cook itself
Click to expand...
I've already tried that, In all fairness even through the fan speed was slower than my profile it made 0 difference to the hotspot but the core temp went up a few C.
 
RichardH754 said:
Yes everything is at stock
Click to expand...
Cheers thanks, I did try Far Cry 6 for a couple of hours yesterday but had the FPS capped at 100 so wasn't using the GPU fully and it was around 80-85c for the hotspot.

I'm beginning to think certain titles push the hotspot further even though the main core temp and wattage is still maxed out.
 
It’s crazy that the GPU hot spot is 101C they not meant to hit 85c isn’t it.

speak to OCUK and get it RMAed tbh

Also make sure you got the latest GPUz software. I think you are using HWinfo64. I would double check those sensor reading with GPUz

If your GPU is hitting 101c there will be throttling
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom