• 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.

NVIDIA 4000 Series

I don't think it's that now because weirdly I've just ran Port Royal demo and benchmark and it completed without a hitch and gave me a result of 26000 and I noticed Port Royal was pulling even more power than Timespy, Timespy max I roughly saw was about 425-430 I think and Port Royal pulled 446w for a lengthy time too so something is odd either with the divers or my Timespy is corrupt somehow.

It's the Focus Gold edition and 3 Yrs old.

There wasn't any corrupt Windows files anyway according to the SFC scan, Still not sure why Timespy won't complete but I may try reinstalling it or try some other GPU tests to see what's going on with that ntddll.dll file.
Maybe it's unstable at the boost clock in TS, guessing port Royal will run the clocks a bit lower due to the RT.

Try running TS but underclocking the core a bit.
 
There wasn't any corrupt Windows files anyway according to the SFC scan, Still not sure why Timespy won't complete but I may try reinstalling it or try some other GPU tests to see what's going on with that ntddll.dll file.

One test to try is set your power limit if using AB to 80% and see if the TS bench completes.
 
Mines the same :mad:

R2tINEW.jpg
WD-40 should work to get that stuff off.

But best try the usual alcohol first.
 
Last edited:
Well that's not good: running Chrome uses 100% GPU with my 4090. If you disable hardware aceleration it fluctuates between 75% and 100%. This also happens if I run Edge.
 
3 gives 100% power, the extra one is just if you're overclocking the power limit and seems like a gimmick when the card is voltage limited anyway.

you don;t get a limited card with 3 of them, the reviewsI looked at arte claiming the Suprim X is 2500mhhz odd boost clock but mine is doing 2800+ on 3x PCIE
Thanks good to know. I have a atx3.0 psu on the way next week so should be ok.
 
I spoke too soon...

Anyone having issues with the Inno3d Ichill X3 with Timespy both normal and Extreme? Mine will not complete a test, Freezes then 3d Mark returns to menu just saying benchmark could not be completed due to error?

Temps are good never went over 66c but he benchmark doesn't last until the second scene then freezes.

Windows 11 22H2
Seasonic 850w Gold PSU using the 3 way adaptor.
12900k on Rog Strix Z690 Gaming A
Latest Nvidia drivers full DDU clean install.

Not sure if I have a duff inno 4090 here.

Edit...

I just noticed everytime it does it there's an error in event log saying Application error 3DMarkTimeSpy.exe faulting module C:/Windows/System32/ntd.dll
check in event logs if there is display drivver crash , is it? I bet card unstable on stock <it happens> But you can rma it and will be fine.
 
Last edited:
Maybe it's unstable at the boost clock in TS, guessing port Royal will run the clocks a bit lower due to the RT.

Try running TS but underclocking the core a bit.
It's kinda weird because now I've ran both Port Royal and Firestrike Ultra and extreme and both passed with correct results, In fact I'm now boosting to 3ghz just to confirm it's not a GPU issue I've added +200 MHz in afertburner and it's passing fine with Port Royal and Firestrike also.

I'm beginning to think my Timespy benchmark is corrupt, I played Far Cry 6 earlier for 40 mins with no issues either.
 
One test to try is set your power limit if using AB to 80% and see if the TS bench completes.
Yeah that was my next test before I've now found out Firestrike and Firestrike Ultra is passing fine and I've now overclocked it to 3ghz to eliminate a clock boost or dodgy chip issue and that passed also, I think my Timespy benchmark is broken somehow, I'll try reinstalling it or maybe wait for next Nvidia drivers.
 
Guys listen. I am hungry to get 4090 and sell 3090 ti. But i wanna ask you about this.

This is new benchmark.


4090 on 1440P have 100fps.
i have on 3090 ti 60fps. Is any sense to cchange for 4090 in 1440P use?
 
Last edited:
One test to try is set your power limit if using AB to 80% and see if the TS bench completes.
Yeah that was my next test before I've now found out Firestrike and Firestrike Ultra is passing fine and I've now overclocked it to 3ghz to eliminate a clock boost or dodgy chip issue and that passed also, I think my Timespy benchmark is broken somehow, I'll try reinstalling it or maybe wait for next Nvidia drivers
check in event logs if there is display drivver crash , is it? I bet card unstable on stock <it happens> But you can rma it and will be fine.
Yeah there's nothing like that related to Nvidia drivers in the event log, I've done some other tests since and overclocked also and it's passing other more strenuous loads so I'm beginning to think my Timespy benchmark is corrupt, I'll try reinstalling it thanks.
 
Yeah that was my next test before I've now found out Firestrike and Firestrike Ultra is passing fine and I've now overclocked it to 3ghz to eliminate a clock boost or dodgy chip issue and that passed also, I think my Timespy benchmark is broken somehow, I'll try reinstalling it or maybe wait for next Nvidia drivers.
Dont worry. I remember in past i had one crash in timespy after installing new drivers. I think it was on 2080 ti. Crash happened once and never again. Maybe something in background?:)
 
Guys listen. I am hungry to get 4090 and sell 3090 ti. But i wanna ask you about this.

This is new benchmark.


4090 on 1440P have 100fps.
i have on 3090 ti 60fps. Is any sense to cchange for 4090 in 1440P use?
You only gain around 25% avg extra performance over a 3090ti at 1440p now if you think that's worth £1800 then go for it.
 
Back
Top Bottom