• 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.

AMD Gimping HBM clock on Fiji after v16.12.2 WHQL driver

So what I reading, FuryX before Crimson using Overdrive and a BIOS Mod allowed Memory Overclocking..
Now that Overdrive is gone and we have Wattman this guy is complaining that his GPU using BIOS mod no longer supports memory overclocking LOL

Now I could totally understand if HBM1 allowed overclocking to begin with and AMD blocked it with Wattman or new driver you defo have a right to complain then...
But the fact you modding the GPU is very wrong to complain, you mod works on old drivers you just after deal with that. If you want new drivers and features then its "Upgrade Time"
 
Puppetmaster said:
What purpose are you using the cards for that the performance matters? Memory overclock tends to not offer great returns in gaming performance (maybe it's different with hbm?).

Are you crypto currency mining?
Click to expand...

On HBM GPUs the HBM speed gives "a lot" of gains in games, it's way more beneficial than core overclocking all without adding much to the TGP
 
gupsterg said:
I have Wattman in v16.12.2, I have performance gain from HBM clock change.

It is from v17.x.x AMD have gimped Fiji.

But I roll over Matt.

If I can't get an AMD rep to agree that AMD have blocked performance from 2017 onwards I got no chance getting it stated further up the chain.

Clearly I need to rethink buying AMD products in the future.

Thank you for your time and I hope I did not upset you. I merely wanted my HW still to perform at it's best going on in the future, like it has done for over a year, clearly I can't have that.

All the best chap. I shall stay mute from now on the subject.

At least you can now breath a sigh of relief.
Click to expand...


Well I got solution for You sell AMD blocked now junk that you got like what 2 or 3 fury x cards?? And buy 1080ti do tdp mod on it and i winning mate :)
 
gupsterg said:
I have Wattman in v16.12.2, I have performance gain from HBM clock change.

It is from v17.x.x AMD have gimped Fiji.

But I roll over Matt.

If I can't get an AMD rep to agree that AMD have blocked performance from 2017 onwards I got no chance getting it stated further up the chain.

Clearly I need to rethink buying AMD products in the future.

Thank you for your time and I hope I did not upset you. I merely wanted my HW still to perform at it's best going on in the future, like it has done for over a year, clearly I can't have that.

All the best chap. I shall stay mute from now on the subject.

At least you can now breath a sigh of relief.
Click to expand...

Overclocking HBM on the Fury X cards in crossfire has never been much fun from day 1.

Wattman was not even around when Fiji launched so it is a bit unfair to expect it to work perfectly.

Even for something like RX480s and mGPU Wattman is a pain in the backside and it is better to use other overclocking software.

As to the recent drivers and mGPU Fury Xs my 4 way setup seems to run really well on them and AMD have done a good job with support.

For gaming Fury Xs are nice cards and you don't have to overclock the memory.
 
His beef isn't with the fact the wattdude doesn't allow to OC HBM, or that 3rd party apps don't allow HBM OCs, his beef is that since the 2017 drivers, any HBM OC that is made directly at the vbios level is discarded, and HBM OC at the vbios level worked correctly with any drivers before 2017, and even worked correctly when wattdude was available on Fiji in the dec 2016 drivers. So he is suggesting that AMD is actively blocking vbios HBM OC at a driver level and that it is not a byproduct of wattdude being implemented.
 
WeSbO said:
His beef isn't with the fact the wattdude doesn't allow to OC HBM, or that 3rd party apps don't allow HBM OCs, his beef is that since the 2017 drivers, any HBM OC that is made directly at the vbios level is discarded, and HBM OC at the vbios level worked correctly with any drivers before 2017, and even worked correctly when wattdude was available on Fiji in the dec 2016 drivers. So he is suggesting that AMD is actively blocking vbios HBM OC at a driver level and that it is not a byproduct of wattdude being implemented.
Click to expand...

Using CF we have always been blocked from overclocking the HBM, IIRC it was only the primary card that would overclock on the memory and the rest would remain at stock. I think there was a longwinded way of getting around this but I never bothered.

Even running in single mode and overclocking the memory it was only worth going to 545mhz on any of my cards, any higher usually resulted in a crash.

I don't even know why the OP is even bothering with this as time has moved on and Vega is here, time for him to sell all those old cards and go for something new.:)
 
Kaapstad said:
Using CF we have always been blocked from overclocking the HBM, IIRC it was only the primary card that would overclock on the memory and the rest would remain at stock. I think there was a longwinded way of getting around this but I never bothered.

Even running in single mode and overclocking the memory it was only worth going to 545mhz on any of my cards, any higher usually resulted in a crash.

I don't even know why the OP is even bothering with this as time has moved on and Vega is here, time for him to sell all those old cards and go for something new.:)
Click to expand...

Not everyone can afford, or wants a new card. As with Creative removing features from cards intentionally, there were a lot of people that didn't want to upgrade their sound cards.
 
So the GPU that AMD marketed as an "overclockers dream" hardly overclocks at all on the core and AMD are now actively blocking all known methods of memory overclocking. AMD are starting to make NVidia look like the good guys. There's got to be a lawsuit in there somewhere. :p
 
Last edited:
pandem0nium said:
To make the change they'd have to for someone to raise a change request, document the requirements. Then one day of a developers time to make the change, maybe longer to write a unit test as well. Then the QA team would have to build it into their test plan for release testing of the next driver.

It's not worth it and they're just not going to do it.
Click to expand...
Can you imagine the request? lol. "It turns out people were hacking the VBIOS on one of our cards to do something with it that was never intended and they're having trouble doing this with the new drivers, can we alter our software to accommodate them?" XD
 
Scougar said:
"never intended". Sure AMD, sure.

You know... not like AMD and NVidia produced tools that overclocked or anything... oh wait...
Click to expand...

You did read the thread right?

AMD didn't give any official way to do this particular overclocking.

So when it stops working ages later in a patch AMD couldn't care less and why should they go back and fix up a new patch to support something they never allowed.

It's not free to find out why it isn't working and then authorise changes to be made.
 
Hotwired said:
You did read the thread right?

AMD didn't give any official way to do this particular overclocking.

So when it stops working ages later in a patch AMD couldn't care less and why should they go back and fix up a new patch to support something they never allowed.

It's not free to find out why it isn't working and then authorise changes to be made.
Click to expand...

No. I can't read. /S

(patiently waits for a mod's warning lol)
 
Hotwired said:
You did read the thread right?

AMD didn't give any official way to do this particular overclocking.

So when it stops working ages later in a patch AMD couldn't care less and why should they go back and fix up a new patch to support something they never allowed.

It's not free to find out why it isn't working and then authorise changes to be made.
Click to expand...
Get out of here with your logical rationale.
 
ubersonic said:
Can you imagine the request? lol. "It turns out people were hacking the VBIOS on one of our cards to do something with it that was never intended and they're having trouble doing this with the new drivers, can we alter our software to accommodate them?" XD
Click to expand...

Hah, yeah. I'm not saying it's unreasonable to ask. It's just that on an answer of no, the reaction should be "ok, fair enough".
 
The thing is clock information is all stored in the BIOS that's gupstergs point he used to be able to overclock via the bios, if AMD's excuse is that they have never actively supported software overclocking of memory then for what reason have they chosen to overrride what is already stored in the BIOS? there's a huge difference between not actively supporting something and going out of your way to stop it altogether.
 
Last edited:
mmj_uk said:
The thing is clock information is all stored in the BIOS that's gupstergs point he used to be able to overclock via the bios, if AMD's excuse is that they have never actively supported software overclocking of memory then for what reason have they chosen to overrride what is already stored in the BIOS? there's a huge difference between not actively supporting something and going out of your way to stop it altogether.
Click to expand...

Get out of here with your logical rationale. :D

EDIT : Just a joke by the way, nobody get excited :D
 
You must log in or register to reply here.
Back
Top Bottom