VOGONS


Reply 20 of 36, by nuvyi

User metadata
Rank Newbie
Rank
Newbie
MikeSG wrote on 2024-03-04, 12:09:

Does the GPU core temp move when you stress the card? Does it get to 80?

There's no way it's hot if the heatsink isn't hot. Paste looks fine

OK, so I applied some new thermal paste and now it shows 66 degrees on the desktop. Apparently, the contact between the heatsink and the chip got better? 🤨
I played Doom 3 on the highest settings for 10 minutes and the maximum temperature was 72 degrees. The heatsink's still as cold as ice.
upd Attached a photo

Attachments

Reply 21 of 36, by rasz_pl

User metadata
Rank l33t
Rank
l33t

there was no thermal diode at this point in time time (early 2000) for measuring the core, its inside the chip, same for all Pentium3 and Athlons/Durons. There must be external sensor for ambient temp somewhere on the pcb far away from main chip. Sadly my diagrams for graphic cards dont go that far back 🙁 oldest I have are some GTXes.

are you screwing that heatsink flush? hold the card to the light and see if there are any gaps and if it isnt hanging on something like ram chip.

Open Source AT&T Globalyst/NCR/FIC 486-GAC-2 proprietary Cache Module reproduction

Reply 22 of 36, by elszgensa

User metadata
Rank Member
Rank
Member

I recommend giving thermal pads a try, at least temporarily. They perform a bit worse than well applied paste - but they are much easier to use, and also take care of height differences between chips, or off kilter heatsinks to a degree.

Reply 23 of 36, by pete8475

User metadata
Rank Oldbie
Rank
Oldbie
nuvyi wrote on 2024-03-04, 16:44:
OK, so I applied some new thermal paste and now it shows 66 degrees on the desktop. Apparently, the contact between the heatsink […]
Show full quote
MikeSG wrote on 2024-03-04, 12:09:

Does the GPU core temp move when you stress the card? Does it get to 80?

There's no way it's hot if the heatsink isn't hot. Paste looks fine

OK, so I applied some new thermal paste and now it shows 66 degrees on the desktop. Apparently, the contact between the heatsink and the chip got better? 🤨
I played Doom 3 on the highest settings for 10 minutes and the maximum temperature was 72 degrees. The heatsink's still as cold as ice.
upd Attached a photo

To me it still looks like that heatsink is still barely contacting any parts of that chip. Again lots of thermal paste is WAY BETTER than air.

Bluetooth heatsinks aren't a thing. 🤣

elszgensa wrote on 2024-03-04, 19:20:

I recommend giving thermal pads a try, at least temporarily. They perform a bit worse than well applied paste - but they are much easier to use, and also take care of height differences between chips, or off kilter heatsinks to a degree.

This is a good idea, a heatsink with a thermal pad is significantly better than a heatsink with paste that doesn't actually touch the chip.

Reply 24 of 36, by iraito

User metadata
Rank Member
Rank
Member

To me it looks like it barely touches, go with the pad idea and if that solves it then you can try some copper shimming for a more effective solution or an aftermarket cooler.

uRj9ajU.pngqZbxQbV.png
If you wanna check a blue ball playing retro PC games
MIDI Devices: RA-50 (modded to MT-32) SC-55

Reply 25 of 36, by Hoping

User metadata
Rank Oldbie
Rank
Oldbie

A curiosity is that many of these chips of the time were slightly convex, perhaps you could try the thermal paste grain of rice method in the center and let the pressure of the heatsink spread it out.
I have the same card, if I have time I will take a look at it and see what it does.
What version of Nvidia drivers did you use?

Reply 26 of 36, by nuvyi

User metadata
Rank Newbie
Rank
Newbie

It's just ridiculous. I added a bit more paste, and now the temps are a little lower, about 61 degrees on the desktop, still a lot.. 😆 OK, I'm going to give the thermal pads a try.

Last edited by nuvyi on 2024-03-04, 21:02. Edited 1 time in total.

Reply 27 of 36, by nuvyi

User metadata
Rank Newbie
Rank
Newbie
Hoping wrote on 2024-03-04, 20:53:

A curiosity is that many of these chips of the time were slightly convex, perhaps you could try the thermal paste grain of rice method in the center and let the pressure of the heatsink spread it out.
I have the same card, if I have time I will take a look at it and see what it does.
What version of Nvidia drivers did you use?

Great. It will be interesting to compare. I'm using version 45.23.

Reply 28 of 36, by nuvyi

User metadata
Rank Newbie
Rank
Newbie
rasz_pl wrote on 2024-03-04, 18:15:

there was no thermal diode at this point in time time (early 2000) for measuring the core, its inside the chip, same for all Pentium3 and Athlons/Durons. There must be external sensor for ambient temp somewhere on the pcb far away from main chip. Sadly my diagrams for graphic cards dont go that far back 🙁 oldest I have are some GTXes.

are you screwing that heatsink flush? hold the card to the light and see if there are any gaps and if it isnt hanging on something like ram chip.

The heatsink is mounted differently.

Attachments

  • 2077_1.jpg
    Filename
    2077_1.jpg
    File size
    156.07 KiB
    Views
    388 views
    File license
    Public domain

Reply 29 of 36, by elszgensa

User metadata
Rank Member
Rank
Member

No screws to tighten, no springs or other tensors that I can see. If that card came from the factory with pads (I have no idea about this model, can't say either way), which are thicker than a layer of paste, you now have less mounting pressure, which certainly doesn't help.

edit: Apparently there's supposed to be a pad on the back too to make good use of that side's sink, don't forget that one. Oh and further down I saw paste in all but one contact point on the sinks so that's alright.

Reply 30 of 36, by The Serpent Rider

User metadata
Rank l33t++
Rank
l33t++

The cooler has a shitty contact with everything. Adding more paste won't really fix the issue, use thermal pads.

I must be some kind of standard: the anonymous gangbanger of the 21st century.

Reply 32 of 36, by Hoping

User metadata
Rank Oldbie
Rank
Oldbie

Well, I think something is wrong with that graphics card, because mine does not heat up at all with the default frequencies, between 43 and 52 degrees, and it is even quite good at overclocking, +100 MHz core and +66 MHz ram, I think it can go even higher, but I only did it to check if it would overheat, but the maximum core temperature was 70 after passing 3DMARk2001 as you can see in the screenshot.

Attachments

  • FX5600.JPG
    Filename
    FX5600.JPG
    File size
    362.67 KiB
    Views
    309 views
    File license
    Fair use/fair dealing exception

Reply 33 of 36, by nuvyi

User metadata
Rank Newbie
Rank
Newbie
Hoping wrote on 2024-03-05, 13:07:

Well, I think something is wrong with that graphics card, because mine does not heat up at all with the default frequencies, between 43 and 52 degrees, and it is even quite good at overclocking, +100 MHz core and +66 MHz ram, I think it can go even higher, but I only did it to check if it would overheat, but the maximum core temperature was 70 after passing 3DMARk2001 as you can see in the screenshot.

Interesting, 52 is on desktop? I mentioned earlier that when I was playing Doom 3 on high settings, temperature got up to 72 degrees max, pretty close. Is your heatsink hot to touch?

Reply 34 of 36, by Socket3

User metadata
Rank Oldbie
Rank
Oldbie
nuvyi wrote on 2024-03-04, 09:11:
kingcake wrote on 2024-03-04, 01:26:

Did you get paste on the components around the die? There's usually a thermal sense diode. If you get thermal paste that is conductive or even capacitive on it, you will get crazy temp readings.

I've had to fix this many times on repaste jobs people did.

One reason I like MX-4, you can smear that all over exposed components and nothing will happen, except for making a mess 🤣

And where's that thermal sense diode? I took a photo, did I get it right? By the way, I used exactly MX4.

I just got one of these (Leadtek A310) and it's cool to the touch. Didn't even get to replace the thermal compound, so far I've just run a few benchmarks, and my IR thermometer reads 39C on the backplate and 41C near one of the mosfets. I didn't check forceware's temp readout, but 70+ for an FX 5600 sounds like a lot.

Reply 35 of 36, by Hoping

User metadata
Rank Oldbie
Rank
Oldbie

I just tried Doom 3 1.1, it doesn't seem very playable to me on this graphics card. We all know that the FX line was garbage in DX9, even at 800x600 and overclocked to 434mhz core and 566mhz vram; It's very slow, pretty sure it's only around 20fps.
Yes, 52 Cº is on the desktop and after playing for about 10 minutes it reaches 64 Cº; overclocked and with the case closed.
The heatsink is the original and the thermal paste is MX4.
The overclock and the closed case are all to try to overheat it, but it doesn't seem to be possible.
When I got this graphics card, the fan was damaged, specifically, the fan PCB was broken due to poor assembly, either at the factory or by the previous owner; It's a silly question, but does the fan work well?

Reply 36 of 36, by nuvyi

User metadata
Rank Newbie
Rank
Newbie
Hoping wrote on 2024-03-05, 18:30:
I just tried Doom 3 1.1, it doesn't seem very playable to me on this graphics card. We all know that the FX line was garbage in […]
Show full quote

I just tried Doom 3 1.1, it doesn't seem very playable to me on this graphics card. We all know that the FX line was garbage in DX9, even at 800x600 and overclocked to 434mhz core and 566mhz vram; It's very slow, pretty sure it's only around 20fps.
Yes, 52 Cº is on the desktop and after playing for about 10 minutes it reaches 64 Cº; overclocked and with the case closed.
The heatsink is the original and the thermal paste is MX4.
The overclock and the closed case are all to try to overheat it, but it doesn't seem to be possible.
When I got this graphics card, the fan was damaged, specifically, the fan PCB was broken due to poor assembly, either at the factory or by the previous owner; It's a silly question, but does the fan work well?

Same cards, but mine shows higher temps yikes. Yeah, I simply picked this game to stress the card. Doom 3 is totally unplayable. The fan is working fine though, but it's kinda noisy.