Notes on modding HD7990's BIOS

Sharing is caring so here's some info of questionable quality on HD 7990 BIOS modding. I haven't tested any of the modifications yet and even it I did the usual applies to BIOS modding. If you kill your stuff due to a mod gone wrong it's not my fault so mod responsibly.

The HD 7990 Buildzoid edition. Since this photo I've replaced all the thermal pads. Changed thermal paste and changed the mounting hardware. I still haven't done the cap mods. In it current state it did this score:

 So it's already pretty fast. However pretty fast is just not enough. Ideally I would like to push the card beyond 1200MHz core maybe all the way to 1300MHz. Now I haven't really tested the limits of this card. Really that 5.9K Heaven score was me taking it "easy" during some late night testing. I expect to hit the card's power limit pretty hard some time soon so I figured I would look into modding the BIOS for MOAR POWER. Since I started digging in the BIOS I figure that I might as well also look into seeing if I could figure out how memory timings on the card work. Here are the notes on that:

HD 7990 reference
Power limit is set with 2 values
Min power
Max power
these are X% apart from the stock TDP. X% is the overdrive power limit slider

EXP
300W BIOS 20% = 360W max and 240W min
300W BIOS 50% = 450W max and 150W min

Power limits are found in this:


D9 00 05 00 E8 03 58 00 00 80 07 00 10 00 00 02 0A 2C 00 00 69 00 DB 00 05 23 01 0A 00 32 01
42 01 3D 03 00 00 C6 16 00 00 58 01 6D 01 73 01 00 00 A1 01 00 00 B3 00 00 00 77 00 00 00 60
EA 00 00 88 01 20 03 00 00 14 00 40

In this stock 7990 BIOS B3(179W) is the Hi limit and 77(119W) the Lo limit. I assume the current limit is also somewhere in that block. If you want to set a value greater than 255W it would look like this:

D9 00 05 00 E8 03 58 00 00 80 07 00 10 00 00 02 0A 2C 00 00 69 00 DB 00 05 23 01 0A 00 32 01
42 01 3D 03 00 00 C6 16 00 00 58 01 6D 01 73 01 00 00 A1 01 00 00 2C 01 00 00 04 01 00 00 60
EA 00 00 88 01 20 03 00 00 14 00 40

This gives you a 300W high limit and a 260W low limit. Which I think will translate into 280W stock.
I find that the original 179W maximum power limit of the reference HD 7990 is kinda concerning especially once you consider that the Vcore VRM for each of the 2 GPUs is only 4 phases. On stock settings the maximum current through them works out to only ~149A at 1.2V. I think 200A is probably safe however without a datasheet for the Volterra power stages I would recommend proceeding with extreme caution while carefully monitoring VRM temperatures(GPU-z supports VRM temps for the ref HD 7990).

MEMORY STUFF

Timing Straps
A timing strap is composed of 48 bytes
98 AB 02 = 02AB98 = 1750Mhz
77 71 33 20 00 00 00 00 31 62 7C 47 80 55 11 11 30 A7 1A 07 00 4C 06 01 22 22 9D 00 6C 0F 14 20 6A 89 00 A0 00 00 01 20 19 12 2F 36 48 28 31 15

C4 7A 02 = 027AC4 = 1625MHz
77 71 33 20 00 00 00 00 10 5A 7B 41 80 55 11 11 2E A5 99 06 00 4C 06 01 22 11 9D 00 6C 0F 14 20 6A 89 00 A0 00 00 01 20 17 11 2B 31 42 26 2F 15

F0 49 02 = 0249F0 = 1500MHz
77 71 33 20 00 00 00 00 CE 51 6A 3B 70 55 10 10 2B A2 18 06 00 4A E6 00 22 00 9D 00 64 0E 14 20 6A 89 00 A0 00 00 01 20 15 0F 27 2D 3C 23 2C 14

1C 19 02 = 02191C = 1375MHz
77 71 33 20 00 00 00 00 AD CD 69 37 70 55 0F 10 29 21 98 05 00 4A E5 00 22 EE 1C 00 64 0D 14 20 5A 89 00 A0 00 00 01 20 14 0E 24 2A 38 22 2A 14

48 E8 01 = 01E848 = 1250MHz
77 71 33 20 00 00 00 00 8C C5 58 31 60 55 0F 0F 25 1E 17 05 00 48 C4 00 22 CC 1C 00 5C 0B 14 20 4A 89 00 A0 00 00 01 20 12 0D 20 25 32 1F 26 13

74 B7 01 = 01B774 = 1125MHz
55 51 33 20 00 00 00 00 6B BD 57 2D 60 55 0D 0E 22 9C 96 04 00 28 C3 00 22 BB 1C 00 53 0A 14 20 BA 88 00 A0 00 00 01 20 10 0C 1E 22 2E 1D 23 12

A0 86 01 = 0186A0 = 1000MHz
55 51 33 20 00 00 00 00 29 B5 46 27 50 55 0C 0D 1E 99 05 04 00 26 A2 00 22 AA 1C 00 4B 08 14 20 AA 88 00 A0 00 00 01 20 0E 0A 1A 1E 28 1A 1F 11

90 5F 01 = 015F90 = 900MHz
55 51 33 20 00 00 00 00 29 31 46 24 50 55 0C 0D 1C 18 A5 03 00 26 A1 00 22 AA 1C 00 4B 07 14 20 9A 88 00 A0 00 00 01 20 0D 0A 18 1B 25 19 1D 11

80 38 01 = 013880 = 800MHz
55 51 33 20 00 00 00 00 E7 AC 35 20 50 55 0B 0D 1A 97 34 03 00 24 81 00 22 AA 1C 00 4B 06 14 20 9A 88 00 A0 00 00 01 20 0C 08 15 19 21 18 1B 11

40 9C 00 = 009C40 = 400MHz
33 31 33 20 00 00 00 00 84 94 22 10 F0 54 09 06 0F 0B A2 01 00 23 80 00 22 AA 1C 00 12 01 14 20 8A 88 00 A0 00 00 01 20 06 05 0B 0C 11 0C 10 0D

Well that's all for this. Hopefully you find this useful for your own BIOS modding. Though to be completely honest I mostly wanted to post this so I would have an easily accessible online backup of my BIOS modding notes because I keep forgetting how I did stuff the last time I modded a certain BIOS. I will probably be posting more notes like this for other cards.

ASUS RX 480 STRIX OC review

Thank you to ASUS for loaning the RX 480 STRIX OC

The backplate adds rigidity and is separate from the main heasink so it's not necessary to remove both if you want to get to the front of the PCB. It does not help cool the PCB but the holes over the VRM section stop it from causing unnecessary heat build up in the VRM area.




Specs
2304 stream processors clocked at 1310mhz
8GB of GDDR5 clocked at 2000mhz on a 256bit bus
1 8pin power connector
Core clock throttling temperature: 85C°
2 HDMI port, 2 DisplayPorts, 1 DVI

Physical Specs
length: 280mm
width: 124mm
height: 2 slots

Cooling

The noise measurements were taken 20cm away from the card on an open air test bench.
The GPU was kept under load by the Graphics Test 1 from 3Dmark Firestrike running windowed using the regular preset windowed at1440x900p with 8xMSAA.
The OC test is 1370Mhz(higher core clocks simply were not stable at high temperatures) core with +50% power limit +100mv using sapphire Trixx.
I test from the lowest fan speed to the highest this means that I actually wait for temperatures to come down this prevents me from getting low temperature measurements by taking measurements too early.

Notes:
The RX 480 Strix I have has a 86% ASIC quality and the core pulls about 20W more than the core of the RX 480 GTR I last reviewed. So even though this card delta temperatures aren't stellar that is mostly due to the power draw discrepancy between the cards.

In the OC test with stock thermal paste the card started throttling at around 86C. This is not due the temperature limit which is at 90C for this card like very other RX 480 I've tested but due to the fact that higher temperatures cause higher power draw and at 86C core temperature that card was pulling well over the 165W power limit that an RX 480 with +50% power has.

An important thing to note is that due to the way the STRIX's heatsink dumps hot air you will want to put some airflow in front of the card's fans or they will end up re-using air that already went through the heatsink which leads to sub optimal thermal performance. If you have good case airflow this should be none issue. If you're not sure I recommend that you try playing with fan placements a little. If you don't want to mess with airflow you'll have to live with the noise of higher fan speed because as far as I know most of the custom 480s have the same problem and the reference card is really really loud if you want get some real OCing done.






The STRIX heatsink is an obvious case of too many fins and too much airflow without enough heat transfer from the core as increasing the fan speed doesn't lead to a very significant drop in temperatures. I suspect this may have something to do with the ASUS direct contact heatpipe design as only 3 heatpipes get proper contact. I went on to try test this theory by attaching thermal probes to the heat pipes and at maximum fan speed there is a 5°C delta between the heat pipe with the most die contact and the outer most heatpipe.

It also worth noting the absolutely massive drop in temperatures going from the stock thermal paste(which judging by the lack of damage to the warranty stickers on the screws was not tampered with) to my current favorite paste Thermal Grizzly Kryonaut. So I highly recommend that you replace the thermal paste on your RX 480 STRIX if the heatsinks seems to under performing. An 8°C temperature drop on RX 480 can make about a 10MHz difference in stable core clock.

AHOC RX 480 Heatsink ranking at a glance:
ASUS RX 480 STRIX
XFX RX 480 GTR
AMD Reference

VRM

Core
- 6 phases controlled by Digi+ ASP1300 (re-branded IR3567B)
- IR3555 PowIRstanges
- Unknown switching frequency
- 360A, 1.5V @ 125°C and 304KHz switching frequency

VRAM
- 1 phase controlled by uP1541R
- 2 low side QM3056 MOSFETs
- 1 high side QM3054 MOSFET
- 40A, 1.5V @ 125°C assuming 500KHz switching frequency
- 67A, 1.5V @ 100°C assuming 500KHz switching frequency

AUX
- 1 phase controlled by uP1541R
- 2 low side QM3056 MOSFETs
- 1 high side QM3054 MOSFET
- 40A, 1.5V @ 125°C assuming 500KHz switching frequency
- 67A, 1.5V @ 100°C assuming 500KHz switching frequency
The RX 480 STRIX ties with the GTR for most powerful Vcore VRM. The GTR achieves it's current rating by using IR Direct FETs. The STRIX with IR3555 PowIRPowerstages. The end result is the same. The VRM on both is ridiculous overkill for an RX 480 which even when pushed into the 1.4V range will only draw around 200A. So the STRIX's Vcore VRM is perfectly good for any kind of overclocking endeavor be that on the card's air cooler, a water cooler or LN2. The VRAM and AUX VRMs are both single phase designs which is the same as you would find on any other RX 480(at least as far as I'm aware). Both are capable of delivering 40A and such are of no concern in any kind of overclocking scenario.

Extras:

0RPM fans in idle 
RGB heatsink lighting
Backplate

Overclocking Tips
The STRIX is compatible with both the Elmor RX 480 Air BIOS and the Elmor RX 480 LN2 BIOS which makes really pushing the card a lot easier as both those BIOSs are signed to work with the AMD drivers and come with a 1.4V and 2V Vcore limit. Which is a significant improvement over the 1.3-1.35V limits that regular RX 480 BIOSs come with. The LN2 BIOS does not have GPU core temperature enabled so do not use it on cards unless you are using LN2.

The STRIX PCB comes with soldering points to mod Vcore Vmem and Vaux on the top edge of the PCB. Vcore control is well supported in software so I wouldn't bother with that one however Vmem and Vaux both lack software support so for those the solder points are very useful. The card does not however have a solder point for the 0.95V VRM so you will have to mod that one the usual way if you want to get rid of the black screens(the card is still running the benchmarks just not outputing to your monitor) that RX 480s get when on running benchmarks on LN2.

Conlusion

Once you replace the thermal paste the card has excellent cooling performance at low fan speeds and noise levels. The VRM quality is also far beyond what any gamer would ever need. So this is certainly a good choice if the card is to spend it's life in your daily system.

The VRM are far beyond what an RX 480 will ever need even on LN2 and as such the card is completely capable of smashing HW high scores sitting on a test bench with an LN2 pot where the heatsink would normally be. The volt mod points are especially a nice touch as they make modify the voltages without software support much easier. The only down side this card has is that it lacks dual BIOS functionality. Which is only really a problem if you intended to do a lot of BIOS modding which carries a high risk of flashing a BIOS that won't allow the card to boot.

Physical Review: XFX RX 480 GTR

Thank you to XFX for giving me the GPU.

The backplate adds rigidity and is separate from the main heasink so it's not necessary to remove both if you want to get to the front of the PCB. It does not help cool the PCB but the holes over the VRM section stop it from causing unnecessary heat build up in the VRM area.

Specs
2304 stream processors clocked at 1338mhz
8GB of GDDR5 clocked at 2000mhz on a 256bit bus
1 8pin power connector
Core clock throttling temperature: 89C°
1 HDMI port, 3 DisplayPorts, 1 DVI

Physical Specs
length: 280mm
width: 124mm
height: 2 slots

Cooling

The noise measurements were taken 20cm away from the card on an open air test bench.
The GPU was kept under load by the Graphics Test 1 from 3Dmark Firestrike running windowed using the regular preset windowed at1440x900p with 8xMSAA.
The OC test is 1420Mhz core with +50% power limit +100mv using sapphire Trixx.
I test from the lowest fan speed to the highest this means that I actually wait for temperatures to come down this prevents me from getting low temperature measurements by taking measurements too early.

Notes:
Both the Kryonaut and stock thermal paste ended up hitting the thermal throttle of the card which is 90°C which is why they both ended up with identical Delta T in the graph.

An important thing to note is that due to the way the GTR's heatsink dumps hot air you will want to put some aiflow in front of the card's fans or they will end up re-using air that already went through the heatsink which leads to sub optimal thermal performance. If you have good case airflow this should be none issue. If you're not sure I recomend that you try playing with fan placemenets a little. If you don't want to mess with airflow you'll have to live with the noise of higher fan speed because as far as I know all the custom 480s have the same problem and the reference card is really really loud if you want get some real OCing done.

I think the heatsink is quiet up to about 55% fan speed then again I have a pretty high tolerance for noise since I like to use 2000RPM fans for pretty much everything.

VRM

Core
- 6 phases controlled by an IR 3567B(running at 304KHz)
- 1 low side IRF6894 MOSFET per phase
- 1 high side IRF6811 MOSFET per phase 
- 200KHz up to 2MHz
- 366A, 1.5V @ 125°C and 304KHz switching frequency
- 505A, 1.5V @ 100°C and 304KHz switching frequency

VRAM
- 1 phase controlled by APW8722A (fixed 600KHz frequency)
- 2 low side SM4373 MOSFET
- 1 high side SM4377 MOSFET
- 36A, 2V @ 125°C and 600KHz switching frequency
- 61A, 2V @ 125°C and 600KHz switching frequency

AUX
- 1 phase controlled by APW8722A (fixed 600KHz frequency)
- 1 low side SM4373 MOSFET
- 1 high side SM4377 MOSFET
- 40A, 1.5V @ 125°C and 600KHz switching frequency
- 67A, 1.5V @ 125°C and 600KHz switching frequency

For the Vcore VRM XFX opted to go with the  the IRF6811 and IRF6894. These are the same FETs you would see in the VRM of a Fury X. The GTR also has 6 phases. Which is the same as a Fury X. The end result is that with the default switching frequency of the 304KHz the XFX RX 480 GTR has the most powerful VRM of any RX 480. With 366A available even at 125°C the GTR is capable of handling pretty much any overclocking you might want to do on it and it will do it without even needing a heatsink on the VRM as long as you give it some airflow.

The VRAM and AUX VRMs are or on par with what you would see on most other RX 480s.The APW8722A has a fixed switching frequency of 600KHz(+/-10%) so both the AUX and VRAM should get rather clean power. Both offer plenty of power as the RX 480's memory system runs on about 35W at stock.

Extras:

Dual BIOS
0RPM fans in idle 
VRAM cooling plate
Hard swap fans

 

Conlusion

I love the RX 480 GTR. It offers a very strong Vcore VRM perfect for any type of overclocking you decided to go for. It's also one of the few RX 480s with a BIOS switch. Which I think is a must have for anyone into BIOS modding(I am). The heatsink may not be the most powerful one I've ever seen however it is perfectly capable for both 24/7 usage and benchmarking runs. My only concern with the card is that the Vcore VRM pulls half it's power from the 12V pins of the PCI-e slot. At stock clocks this is not an issue however when you really push an RX 480 the power draw goes through the roof and that could lead to a burn out 24 pin connector if you are running several RX 480 GTRs. But that's really only an issue for me since I'm toying with idea of doing a 4 way RX 480 benching session.

If you like the content you see on AHOC please donate to my Patreon so that I can make more/better content.

The cheapest LN2 GPU pot ever.

GPU pots are expensive. Like really really expensive. Slims go for 100-200$ and can't handle any of the big GPUs high power GPUs. Fat pots are even more expensive ranging from 200-400$. So basically that means I can't run LN2 on GPUs............
......
.......or does it?

Ladies and gentlemen I present you the Tapester GPU LN2 pot.

It's not pretty but it doesn't leak LN2 and it should be enough for some hilarity on a live stream. So then what is the Tapester? It's a reference HD 4870 heatsink wrapped in a liberal amount of electrical tape. It will mount on 4800 and 5800 cards. It should also fit several other AMD cards but I don't have them so I neither care nor know.


Here's some pics of the first iteration with a few hundred ml of LN2 in it.































The surprisingly leak proof bottom


Pouring LN2 into this first version was a very wasteful affair since I didn't give the top much of a funnel into which to pour.

Also during the whole cooling process the pot was making some horrible cracking noises as the tape shrunk from the cold. One of the cracks even shook the whole pot


I let the pot defrost and dry before I did an inspection of the damage that the LN2 did to it.



This is the only place where I found a visible crack in the tape. I think this is the one that shook the pot when it cracked. I'm also pretty sure that the tape cracked because I stretched it too much when putting it on. Either way I'll just add more tape.

The bottom of the HD 4870 has a layer of plastic. The tape didn't stay stuck to this when I added LN2 so this are got an extra piece of tape to hold it together better.

This wasn't caused by the LN2 but by lazy tape placement. one thing I noticed is that the tape held best when it didn't have any loose edges so this had to go.

Here too the tape started coming off. The solution is obviously more tape.

































I added some extra tape to the top for a proper funnel. After all if I can't get LN2 in the pot it isn't a good pot.

Here's the fixed bottom. The tape is just barely low enough to clear the center shim.

Here's a full frontal shot of the abomination.

And here it is fully insulated with 5mm armaflex and ready to roll. To test it I plan to just run some Unigine Heaven Basic preset with the 4790K on air cooling. If this works I'll have a few more HWpoints to add to my total. But much more importantly I will a working LN2 pot for 2 of my GPUs.Unfortunately I won't do much benching with this for a while because to get a good amount of points I'd also need LN2 on the CPU and I currently don't have the equipment needed to LN2 both a GPU and a CPU.

If you enjoy the content you see on Actually Hardcore Overclocking and would like to see more please share it or donate using the donate button in the top left of the site. All donations will be used to make more content even if just helps buy more themal paste. Right now though all donations will go towards a capture card.

Sapphire R9 Fury Tri-X OC Physical review.

It's hard to see but Sapphire put a small piece of plastic over the interposer to protect it. It isn't perfect since you can still break the interposer but much better than nothing.

Buildzoid's GPU reviews explained

Specs
3584 stream processors clocked at 1040mhz
4GB of HBM clocked at 500mhz on a 4096bit bus
2 8pin power connectors officially supporting up to 375W
Core clock throttling temperature: 85C°
1 HDMI port3 DisplayPort

Physical Specs
length: 320mm
width: 110mm
height: 2.5 slots

Cooling: 10/10

 When I first played a game on this card I knew I was gonna end up giving away points like they grow on trees. And why wouldn't I the cooling on this card can get the same temperatures as a custom water block. I currently don't have the equipment necessary to do proper noise testing but just ot give you an idea of the fan noise I'll tell you this. With the GPU at 0RPM my sound meter reads about 51.5dB sitting 8cm from the GPU. With the fans at 40% it reads 52 dB. At 55% it gets 59dB. At 75% it reads 66dB and at 100% it reads 73dB. I'd recommend running the card around 55-65% fan speed since that's when the noise temperature trade off is quite reasonable.

VRM: 9/10
Core
- 6 true phases provided by an IR 3567B
- 1 low side IR 6894 MOSFETs providing 105A per phase at 100C case temperature
- 1 high side IR 6811 MOSFET providing 44A per phase at 100C case temperature
- 200KHz up to 2MHz
VRAM
- 1 phase
- 1 low side IR 6894 MOSFETs providing 105A per phase at 100C case temperature
- 1 high side IR 6811 MOSFET providing 44A per phase at 100C case temperature
VAUX
- 1 phase
- 1 low side IR 6894 MOSFETs providing 105A per phase at 100C case temperature
- 1 high side IR 6811 MOSFET providing 44A per phase at 100C case temperature
The low side on this VRM is a beast. However like on the R9 290X the high side here is iffy. If you keep it cool you'll be fine but if you let the VRM get hot you can't let the card pull much more than 500W because you'll be risking burning out the high side. There are plenty of cases of people doing this on the R9 290X. The Fury uses much less power so it's not as much of a risk but keep it in mind if you're going heavy on the voltage and not using LN2 because the VRM can easily overheat. Also do note that in certain situations the card has coil whine. Not enough to be as annoying as say loud fans but enough to be noticeable so keep that in mind. I will try somethings with the card to get rid of the coil whine and will put that into my Fury volt modding post which is coming soon.

Extras:
+1 Dual BIOS
+1 0RPM fans in idle 
+1 VRAM cooling
+1 Backplate
+1 Silicon Interposer protector

Overclocking
The highest I've managed to push this card to was 1144/566mhz using Sapphire Trixx. These were not 100% stable clocks. My 100% stable clocks are 1100/550mhz and require a custom fan curve to keep the GPU sub 55C core temperature. The GPU does not clock well and instabilities are hard to catch. I had the card running perfectly OK at 1135mhz core clock for several hours and then it black screened out of no where when playing a relatively light game. This leads me to believe that the dynamic power management is holding the card back.

Conclusion: 24/20 Why is the score so high?
I love this card. It is by far the best GPU I have ever had. It is way way quieter and cooler than all my past GPUs. It's so good a card that I'm terrified of volt modding it(also because the 3567B has the weirdest voltage sensing circuit I've ever dealt with). Which I'm working on as I finish up this review. I don't really have much to say. I love this card it provides an excellent platform for volt modded overclocking.

The voltmod guide for this GPU is now up

Thank you to CoolerMaster for providing the V1000 PSU that was used for this review.

The road to voltage Fury.

So I just made major step towards making the R9 Fury volt mod guide. I have revived the R9 290X Windforce which I killed and managed to get it's VRM under manual control.

Now most of you might be sitting there going DAFUQ How does volt modding an R9 290X have anything to do with volt modding the Fury?
It's simple. The Fury and R9 290X use the same exact voltage controller. If I manage to control the IR 3567B on the 290X it makes taking control of the IR 3567B on the Fury that much simpler.

The first thing I learned about the IR 3567B is that it's pretty smart. So smart, that the damn thing will not let the card run if it doesn't sense current on the VRM phases. With most GPU voltage controllers you cut the power sensing lines and they just ignore the fact that the current reading of 0.8A is wrong and continue running the VRM like nothing's changed. Not the IR 3567B, no the IR 3567B shuts the card down if it sense 0A on the phases. So after I tore all the SMD 300ohm ISEN resistors off the card in hopes that I would remove all power limits. I ended up resoldering every single one of the damn things. However I don't have SMD 300ohm resistors and so I improvised to end up with an R9 290X that looks more like a modern art piece than a GPU.

I also managed to under volt mod the card so it ended up running at 0.56V idle instead of the 0.968V that it should idle and boot at and so naturally it didn't work. I managed to fix that too and the card now boots and runs. At least on stock voltage I haven't tried overvolting it yet because the Windforce cooler no longer fits and I've run out of Seidons to ziptie on to it(it might sound like a terrible idea but the Seidon 120V rev2.0 can keep the 290X sub 70C even with 1.4V and at 55C when running stock volts and clocks).
Either way after this volt modding adventure modding the Fury will be a walk in the park.... I hope.

PS I'm very much aware that the AMD 125W CPU heatsink can't cool an R9 290X however I needed a cooler that could keep the GPU running long enough for me to get to the BIOS and it did that so you can keep the comments about me being an idiot to yourselves thank you very much.


Thank you to Cooler Master for powering this blog by giving me a V1000 PSU.