Overclock.net › Forums › Graphics Cards › AMD/ATI › Software for R9 285 bios edit
New Posts  All Forums:Forum Nav:

Software for R9 285 bios edit - Page 71

post #701 of 1079
@Blackfirehawk

BlackFireHawk_ASUS_R9_380_STRIX_1080Mhz.zip 101k .zip file

BlackFireHawk_ASUS_R9_380_STRIX_1133Mhz.zip 101k .zip file

This is what I have as my general usage BIOS, one for gaming and efficiency at maximum fps at 4k & 5k.

BIOS stock max voltage is 1306mv, which has a TDP of as high as 240 watts... So less voltage is okay actually because each voltage has a TDP limit. I find 1325mv, just 19mv over stock maximum is fine within a 225W maximum deliverable limit. I'm testing TDP 100 Watts/ Stock TDC Amps Current/ 170 Watts maximum deliverable for lower temperatures.

There does come a point when you sit back realize what kind of power the 380X is sucking down, and compare it to faster more efficient less tdp cards. So that's why its key to find the precise lowest stable voltage a certain clock wants to use in All Load Scenarios, then define the TDP accordingly. That way you can make the most efficient use of power to deliver the maximum most economical frames to the display.

From my experience, I like the 380x it's not bad. It's cooler and fairly efficient compared to my 45 Giga pixel/s GTX 580 even know the 580 waxs the 380x in certain pixel intense scenes. The 380x does stuff the gtx 580 couldn't dream of... Like 4k & 5k at Very High & Ultra settings... haha I feel my 380x is right up with the 970 & GTX 1060 among others... I was surprised how much I got waxed by the mighty GTX 770, haha I was like what what?!? So I'm like idk I might wanna go Nvidia Again but idk... I have Two R9 290Xs that Im getting all the caps replaced on atm... getting the DVI ports unsoldered to allow best airflow and getting Vapor-X coolers for them too... Pretty excited that one R9 290X outruns a GTX Titan 6GB with it's 64 ROPs and 512-Bit ram just epic. I feel after I dial in the physical cooling flaws, I'll be able to see what the R9 290X is all about unlike reviews who don't re-engineer the thermals like myself correcting mistakes made which are many.

The card actually works way better at say 1185mv core and just 1000Mhz with a TDP of 170 watts can utilize the full 170 watts to let the 1000 Mega Hertz core clock stretch its legs a bit. Very solid extra smooth ultra low input latency zero or near absolute zero at 1Ghz core clock. Higher clock frequencies call on a bit of memory/ processing command latency delay to deliver more data relatively error free.


Edited by chris89 - 9/23/16 at 8:01am
post #702 of 1079
This must be some kind of Crysis Glitch..... 638FPS at 4K ... DX9 Mode all high hahaha wheee.gif


Edited by chris89 - 9/23/16 at 4:22pm
post #703 of 1079
HAHA Yep I'm back benchmarking the 1172Mhz BIOS I'm confident enough for continuous stability.

I tested GTA V @ 1080p, 1440p, and 2160p with all "High" Settings.

75.47fps @ 380X @ 1080P
Frames Per Second (Higher is better) Min, Max, Avg
Pass 0, 25.519987, 77.460983, 62.648022
Pass 1, 55.551723, 85.632240, 72.433022
Pass 2, 52.177444, 108.642990, 76.191673
Pass 3, 61.993557, 133.397736, 90.106003
Pass 4, 31.492060, 123.615402, 75.985359
72.82fps @ 380X @ 1440P
Frames Per Second (Higher is better) Min, Max, Avg
Pass 0, 28.118008, 76.021477, 61.782478
Pass 1, 51.384766, 85.446335, 72.881676
Pass 2, 53.443077, 99.041260, 73.136826
Pass 3, 50.768150, 119.043770, 82.183403
Pass 4, 30.318445, 117.323250, 74.131058
43.43fps @ 380X @ 2160P
Frames Per Second (Higher is better) Min, Max, Avg
Pass 0, 11.324775, 65.100845, 42.346260
Pass 1, 25.265814, 81.564125, 42.861576
Pass 2, 22.296694, 90.030640, 42.548077
Pass 3, 23.033545, 90.712883, 44.806362
Pass 4, 20.647438, 103.587112, 44.615463




Edited by chris89 - 9/23/16 at 5:46pm
post #704 of 1079
@Smartrior

I started only using the Base Voltage and Zero'ing all the other CAC Fields which are not needed at all.

Here ya go Smartrior. thumb.gif



Smartrior_BIOS.zip 519k .zip file

1)225watt power limit
2)1150Mhz core
3)1550Mhz memory
4) 1331 mv core MAX voltage
5)1175 mv memory MAX voltage
6)fan curve 1)0%-50c 2)70%-70c 3)100%-80c
7) 800mv when idle
post #705 of 1079
@Mynm

You might try to just install the display driver alone in device manager after a DDU...

I'm testing out 16.7.3 WHQL driver alone and it works quite well. Seems to behave better.

Slightly less performance than 16.9.2 but is more stable and has better behavior.
post #706 of 1079
Testing out new thermal stuff and designed a wind tunnel for the gpu in my precision t7500. Moves the air out of the case including the gpu heat pumping into the case. I modded the gpu cooling fan to 5v, which has incredible airflow and acceptable noise, and the hdd fan to 12v direct and the hdd's stay 15C cooler than before.

I've found the true stable everyday gamer clock at 1235Mhz with 180 watt tdp, 140 amps, 225 watt max deliverable. 1235mhz is not stable at 1150mv offset, nor 1175, nor 1200 because it all has to do with what the offset is. Once you go past 1200mhz, 1175 offset isn't stable so the offset needs to be roughly no more than -255mv which is 00 FF...

I still couldn't catch my old 3-way sli gtx 470s on Monster Hunter Online Benchmark. I do think the R9 380X does very well though for what it is. It can do stuff the 3-way sli 470s couldn't dream of resolution wise, haha.









Edited by chris89 - 9/29/16 at 6:35am
post #707 of 1079
I'm getting further to developing a bios with max power savings, along with minimal power consumption during decode.

I find that true stability is always a bit more than what is found to be stable. For instance you don't want to clock as high as a voltage will hold, but just a bit less clock.

We want a bit of overhead voltage.

I found State 0, and 1 have Vddci Offset disabled so I enabled an offset on State 0 & 1. All states 0 to 6 (State 7 is max 3d, so 1125mv offset minimum is possible) can take no more than -25mv or -26 E6 FF... CD FF is -50mv or -51.

Default Values Below
State 0 - A665 - ucCKSVOffsetandDisable - 80 for -128
State 1 - A670 - ucCKSVOffsetandDisable - 80 for -128

New Changed Values Below (Which allows offset in State 0 & 1)
State 0 - A665 - ucCKSVOffsetandDisable - 00 for 0
State 1 - A670 - ucCKSVOffsetandDisable - 00 for 0

(Before)
voltage lut entry 1 - b50a - 4C 04 - 1100mv
(After)
voltage lut entry 1 - b50a - E8 03 - 1000mv

(Before)
A6DD for usVDDCI is 1100 4C 04
A6E1 for usMVDD is 1000 E8 03

(After)
A6DD for usVDDCI is 1000 E8 03
A6E1 for usMVDD is 1000 E8 03

(Default Values) object_3_pcie
a7d7 : usPCIEGenSpeed : 00 : 0
a7d8 : usPCIELaneWidth : 01 : 1
object_2_pcie
a7db : usPCIEGenSpeed : 00 : 0
a7dc : usPCIELaneWidth : 02 : 2
object_1_pcie
a7df : usPCIEGenSpeed : 01 : 1
a7e0 : usPCIELaneWidth : 02 : 2

(Changed New Values) object_3_pcie (Keeps power consumption as low as possible while not in max 3d mode)
a7d7 : usPCIEGenSpeed : 00 : 0
a7d8 : usPCIELaneWidth : 00 : 0
object_2_pcie
a7db : usPCIEGenSpeed : 00 : 0
a7dc : usPCIELaneWidth : 00 : 0
object_1_pcie
a7df : usPCIEGenSpeed : 01 : 1
a7e0 : usPCIELaneWidth : 02 : 2

The Below E6 FF Values are default and unchanged, so values have been adjusted to reflect the offset.

Tier 6 BIOS Power Consumption Tuning
A65C - (Impossible)300Mhz - 700mv - 650mv offset = -51 = CD FF
A65C - (Possible)300Mhz - 700mv - 675mv offset = -26 = E6 FF
400Mhz - 900mv - 875mv offset = -26 = E6 FF
500Mhz - 900mv - 875mv offset = - 26 = E6 FF
600Mhz - 925mv - 900mv offset = -26 = E6 FF
700Mhz - 950mv - 925mv offset = - 26 = E6 FF
800Mhz - 975mv = 950mv offset = -26 = E6 FF
900Mhz - 1000mv - 975mv offset = -26 = E6 FF
1157Mhz - 1350mv - 1125mv offset = -226 = 1E FF



Edited by chris89 - 10/6/16 at 10:29am
post #708 of 1079
Quote:
Originally Posted by chris89 View Post

@Mynm

You might try to just install the display driver alone in device manager after a DDU...

I'm testing out 16.7.3 WHQL driver alone and it works quite well. Seems to behave better.

Slightly less performance than 16.9.2 but is more stable and has better behavior.

I have tested ubuntu and I don't have freezes with videos, so I think is a software problem. I'm testing to use windows driver for HDMI audio instead of amd driver, and disabling it at play black devices to see if it is the problem.
Quote:
Originally Posted by chris89 View Post

Warning: Spoiler! (Click to show)
I'm getting further to developing a bios with max power savings, along with minimal power consumption during decode.

I find that true stability is always a bit more than what is found to be stable. For instance you don't want to clock as high as a voltage will hold, but just a bit less clock.

We want a bit of overhead voltage.

I found State 0, and 1 have Vddci Offset disabled so I enabled an offset on State 0 & 1. All states 0 to 6 (State 7 is max 3d, so 1125mv offset minimum is possible) can take no more than -25mv or -26 E6 FF... CD FF is -50mv or -51.

Default Values Below
State 0 - A665 - ucCKSVOffsetandDisable - 80 for -128
State 1 - A670 - ucCKSVOffsetandDisable - 80 for -128

New Changed Values Below (Which allows offset in State 0 & 1)
State 0 - A665 - ucCKSVOffsetandDisable - 00 for 0
State 1 - A670 - ucCKSVOffsetandDisable - 00 for 0

(Before)
voltage lut entry 1 - b50a - 4C 04 - 1100mv
(After)
voltage lut entry 1 - b50a - E8 03 - 1000mv

(Before)
A6DD for usVDDCI is 1100 4C 04
A6E1 for usMVDD is 1000 E8 03

(After)
A6DD for usVDDCI is 1000 E8 03
A6E1 for usMVDD is 1000 E8 03

(Default Values) object_3_pcie
a7d7 : usPCIEGenSpeed : 00 : 0
a7d8 : usPCIELaneWidth : 01 : 1
object_2_pcie
a7db : usPCIEGenSpeed : 00 : 0
a7dc : usPCIELaneWidth : 02 : 2
object_1_pcie
a7df : usPCIEGenSpeed : 01 : 1
a7e0 : usPCIELaneWidth : 02 : 2

(Changed New Values) object_3_pcie (Keeps power consumption as low as possible while not in max 3d mode)
a7d7 : usPCIEGenSpeed : 00 : 0
a7d8 : usPCIELaneWidth : 00 : 0
object_2_pcie
a7db : usPCIEGenSpeed : 00 : 0
a7dc : usPCIELaneWidth : 00 : 0
object_1_pcie
a7df : usPCIEGenSpeed : 01 : 1
a7e0 : usPCIELaneWidth : 02 : 2

The Below E6 FF Values are default and unchanged, so values have been adjusted to reflect the offset.

Tier 6 BIOS Power Consumption Tuning
A65C - (Impossible)300Mhz - 700mv - 650mv offset = -51 = CD FF
A65C - (Possible)300Mhz - 700mv - 675mv offset = -26 = E6 FF
400Mhz - 900mv - 875mv offset = -26 = E6 FF
500Mhz - 900mv - 875mv offset = - 26 = E6 FF
600Mhz - 925mv - 900mv offset = -26 = E6 FF
700Mhz - 950mv - 925mv offset = - 26 = E6 FF
800Mhz - 975mv = 950mv offset = -26 = E6 FF
900Mhz - 1000mv - 975mv offset = -26 = E6 FF
1157Mhz - 1350mv - 1125mv offset = -226 = 1E FF



Interesting.

Is voltage lut entry 1 - b50a - E8 03 - 1000mv reducing vddci voltage?

Are you blocking dpms on that hwinfo screenshot?
post #709 of 1079
@Mynm

Sorry to hear your still having freezing. I have freezing if any one of the many voltages are too low. It can be stable at first, but later by just scrolling a window down or resizing a video window it will freeze. Always solved with 25 extra millivolts, though you would need to know which dpm voltage is too low that is... Could be any one of the dpm state 0 to 7...

The only difference from Lut Entry Voltage 1 from 1100mv to 1000mv is pretty much no difference and it didn't affect the Core voltage or VDDCI... The card just runs about 1 to 2C cooler... I also like to run the absolute minimum voltage on the offset for dpm 7 among all other dpm states. I find 1125mv on the offset yields lower temperature, less power consumption and ever so slightly more stability. Though with all the reductions, including the usVddci, and usMvdd to both 1000mv yielded another 1 to 2C cooler so totaling about 4 to 5C cooler at all times and no reduction in stability, it seems more stable actually. I also like that even with the MCLK_Dependency usVddci and usMvdd at 1000mv can still hold 1563Mhz memory just fine.

What are your voltages and clocks at the moment? On stock BIOS? To know for sure I would set the clocks far less on stock bios voltage. Basically set DPM 1 to 400mhz rather than 473mhz, then 500mhz, 600mhz, 700mhz, 800mhz, 900mhz, 1000mhz. Since 1000mhz is extremely stable at stock 65288 which is 1.225v/ 1.175v. Does that makes sense? Would you like for me to make you an ultra stable bios with a load of voltage overhead for each clock to determine what it is that is causing the issue?

As far as the screenshots go, I'm using all the dpm states. Also the high temperature there in the screenshot is seconds after heavy 3d load testing and temperatures were still coming down. I have successfully reduced decode from 20 Watts on the core with stock bios, to no more than 10 Watts... While maintaining all the states.

I modded my card a bit more. I unsoldered the DVI Ports. Which resulted in a 30 to 40% fan speed reduction to 10 to 20% reduction in core temperature. Since most of the heat was dumping inside the case because of the exhaust vent being blocked by over 70% with the DVI ports and the mounting plate design. I have a modest fan profile and now no matter how high I clock it or volt it, I see no more than 55% fan speed at 75C. Before would dump loads of heat into the case, minimal air out the exhaust and the fan would go to near 100% to hit upwards of 80C.



Edited by chris89 - 10/6/16 at 6:46pm
post #710 of 1079
Alright guys, I see this thread is still active... I recently put my Powercolor R9 380 4gb under water. Ek r9 285 block and backplate. I was running AB assisted overclocks ranging from 1080/1500-1120/1600 depending on ambient temps and those would leave me around 65c MAX temp under load with 80% fan profile. Now I'm hitting 45c under load at the 1120mhz core clock and interested in upping core voltage and really seeing what this card can do. After reading the majority of this thread I see a number of different approaches to achieving this. My question is which would be easiest/most effective? Will increasing the power limit via AB have the most effect in my reaching a stable 1200+mhz core? Or will a bios mod be required? I'm comfortable doing CPU overclocks, currently running my 9370 daily at 4.84ghz on 'stock' vid and have benched over 5ghz in the 1.5v+ range, but getting more voltage to the GPU is new for me. Any insight appreciated!

Edit: I saw somewhere that the epilida ram doesn't clock well past 1550, which would explain the Mem errors I would get when at 1600. Would a slight voltage boost help here as well or would it be best to just run 1550-1575(I get virtually no errors when running 1550)

Edit 2: Fairly certain tonight after work I will try the registry hack to turn the power limit up via AB and see what that does for me. Really I just have 20+C of thermal headroom just to get back to the operating temps I had under air, so I'd like to try and take advantage of at least some of this.

Edit 3: Figured out the Bios Edit, benching at 1200mhz/1550mhz with a bios VID increase. Still massive thermal headroom running in the 1.3v+ range thus far. Will submit fresh screens shortly.






Edited by m42BMW - 10/8/16 at 3:37pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD/ATI
Overclock.net › Forums › Graphics Cards › AMD/ATI › Software for R9 285 bios edit