Software for R9 285 bios edit - Page 111 - Overclock.net - An Overclocking Community

Forum Jump: 

Software for R9 285 bios edit

Reply
 
Thread Tools
post #1101 of 1125 (permalink) Old 12-26-2018, 07:01 PM
New to Overclock.net
 
mynm's Avatar
 
Join Date: Jan 2016
Posts: 623
Rep: 54 (Unique: 20)
Quote: Originally Posted by BLUuuE View Post
Yes, VDDCI in HWINFO shows as 1.1v and usVddcOffset affects it.

Code:
"F:\MSI Afterburner\MSIAfterburner.exe" /wi6,30,8E,4
Running that also changes the VDDCI in HWINFO.

For the VOI, I tried

Code:
04 00 24 00 00 04 00 00 00 80 10 00 00 00 00 00 39 03 00 00 10 00 B6 03 00 80 00 00 CF 03 00 80 10 00 CF 03
and

Code:
04 00 24 00 00 04 00 00 00 80 10 00 00 00 00 00 39 03 00 00 10 00 B6 03 00 80 00 00 CF 03 00 80 00 00 4C 04
Both didn't affect VDDCI.
This is getting weird, I will try to see your bios tomorrow, your gpu seems to be using usVddcOffset to set the real vddci and not one core phase. Try to not to get more than 1.1v for the vddci.
mynm is offline  
Sponsored Links
Advertisement
 
post #1102 of 1125 (permalink) Old 12-26-2018, 07:08 PM
@integralfx
 
BLUuuE's Avatar
 
Join Date: Feb 2014
Posts: 195
Rep: 7 (Unique: 6)
Quote: Originally Posted by mynm View Post
This is getting weird, I will try to see your bios tomorrow, your gpu seems to be using usVddcOffset to set the real vddci and not one core phase. Try to not to get more than 1.1v for the vddci.
Sounds good.

I also seem to be getting VDDC and VDDCI spikes. VDDCI spikes even happen on the stock BIOS, but I don't think they happen with VDDC. I think it's because I set the LLS to 0 which is causing the VDDC spikes.



Thanks again.

BLUuuE is offline  
post #1103 of 1125 (permalink) Old 12-27-2018, 02:36 AM
New to Overclock.net
 
mynm's Avatar
 
Join Date: Jan 2016
Posts: 623
Rep: 54 (Unique: 20)
Quote: Originally Posted by BLUuuE View Post
Sounds good.

I also seem to be getting VDDC and VDDCI spikes. VDDCI spikes even happen on the stock BIOS, but I don't think they happen with VDDC. I think it's because I set the LLS to 0 which is causing the VDDC spikes.



Thanks again.
vddci LLS for my 380 was independent of the vddc one, I have tested it whit the multimeter and couldn't be modified for my gpu. For your voltage controller is also independent, if fou don't have changed vddci LLS it is the stock value, and maybe you could change it.

Some polaris users also are seeing voltage spikes: https://www.overclock.net/forum/67-a...l#post27046025 as you can see in the link, @hellm , did a mod to the VOI to set an Vmax and was working to aboid spikes.

I also see voltage spikes with LLS1, and also with stock bios and LLS3, but LLS0 is disabling the LLC so you could see more.

The think about the LLS is that hwinfo vddc readings seens to be wrong at least for ncp 81022 voltage controllers. Acording to my multimeter readings, LLS3 had a low vdroop of 10mv and hwinfo was reading 50mv vdroop; with LLS1 was increasing the volage with load but hwinfo was reading a 20mv vdroop.
But I don't know if this hwinfo reading error is the same for your voltage controller, and the only way to know it is to measure it with a multimeter, and is risky.

Edited:

I take a look to your bios and your hwinfo vddci readings seems to be affected by the vddcOffsets like you say, as you can see at the attached image.

But I have no idea if it is the real vddci or not, I have to suppose is not the vddci, because if not your bios is bad configured, and because VOI vddci values have not influence in hwinfo vddci readings.

I was confused because hwinfo vddci values are simirar at the ones of the VOI. And because of the Buildzoid 480 video.
Maybe the 480 is using a 6+1 config whit the vddci into the second loop, and your 380 is usung 5+1 with one of the core phases into the second loop, but no idea why.

But again, I can not know if it is vddci or not without a multimeter measure, and doing it is risky, so better to not to do it.

Edited:

About the voltage peaks my new 580 nitro+ is getting peaks of 1.219v for the vddc, while the stock voltage is 1.15v from dpm 4 to 7, like you can see in the attached image. I have no idea if these readings are errors or what is happening, but of courese I am not going to use the multimeter to test the voltage.

Edited: Into the powerplay image the vddgfxoffsets are vddcoffsets and vddcoffsets vddgfxoffsets, sorry.
Attached Thumbnails
Click image for larger version

Name:	380 asus pwp.png
Views:	17
Size:	80.3 KB
ID:	241600  

Click image for larger version

Name:	voltage peaks.png
Views:	16
Size:	110.6 KB
ID:	241616  


Last edited by mynm; 12-30-2018 at 04:19 AM.
mynm is offline  
Sponsored Links
Advertisement
 
post #1104 of 1125 (permalink) Old 12-27-2018, 04:05 PM
@integralfx
 
BLUuuE's Avatar
 
Join Date: Feb 2014
Posts: 195
Rep: 7 (Unique: 6)
Spoiler!


Thanks for the info.

I'll try setting the Vmax register to 0x43 (0100 0011: 1.250v VDDC 1.138v VDDCI max) and see if it reduces the voltage spikes.

This is what my VOI looks like now.

Code:
62 00 03 01 01 07 0C 00 0E 00 00 00 00 00 00 00 04 00 24 00 00 04 00 00 00 80 10 00 00 00 00 00 39 03 00 00 10 00 B6 03 00 80 00 00 CF 03 00 80 10 00 4C 04 05 03 22 00 08 96 60 00 00 00 00 00 14 00 2A 00 3D 00 43 00 8E 00 00 00 22 00 78 00 23 00 78 00 FF 00 05 07 0C 00 00 00 00 00 00 00 00 00
I manually set a +75mV VDDC offset in MSI Afterburner and +50mV VDDCI offset via

Code:
"F:\MSI Afterburner\MSIAfterburner.exe" /wi6,30,8E,8
but VDDC and VDDCI seem to ignore the Vmax set in BIOS?

I tried reading register 0x3D with
Code:
"F:\MSI Afterburner\MSIAfterburner.exe" /ri6,30,3D
but it reads 0xAA?



I also tried setting Vmax to 0x43 with
Code:
"F:\MSI Afterburner\MSIAfterburner.exe" /wi6,30,3D,43
but after reading it again, it still shows 0xAA.

I think Vmax is working correctly?



The VRM out voltages are still really high, but VDDC and VDDCI don't spike.
Attached Images
 


Last edited by BLUuuE; 12-28-2018 at 01:02 AM.
BLUuuE is offline  
post #1105 of 1125 (permalink) Old 12-28-2018, 01:36 AM
New to Overclock.net
 
mynm's Avatar
 
Join Date: Jan 2016
Posts: 623
Rep: 54 (Unique: 20)
Quote: Originally Posted by BLUuuE View Post
Spoiler!
See this post: https://www.overclock.net/forum/27093361-post607.html, @hellm and @Vento041 found that to activate the Vmax you have to use other register, E4 0 before and E4 to 1 after the 3D register.
@MixedC confirmed here that is working with vrmtool: https://www.overclock.net/forum/27096897-post612.html so it could work with afterburner.


Edited:
I found a higher resolucion image of you pcd: https://www.ixbt.com/video3/images/a...scan-front.jpg

In my attached image you can see the two vrms for the vddci and the memory, I see two littel chips highlighted in green that could be controlling them, so maybe the hwinfo vddci is only one of the core phases. But I can not confirm this because I the name of the chips can't be seen.

Maybe somebody here could hepl you to know if these chips are controlling the vddci and memory vrms, I can not do more. The other way to know the chips name is to open your gpu and see it.
Attached Thumbnails
Click image for larger version

Name:	asus-r9-380-scan-front.jpg
Views:	4
Size:	860.5 KB
ID:	241786  


Last edited by mynm; 12-28-2018 at 01:48 AM.
mynm is offline  
post #1106 of 1125 (permalink) Old 12-28-2018, 04:20 AM
@integralfx
 
BLUuuE's Avatar
 
Join Date: Feb 2014
Posts: 195
Rep: 7 (Unique: 6)
Quote: Originally Posted by mynm View Post
See this post: https://www.overclock.net/forum/27093361-post607.html, @hellm and @Vento041 found that to activate the Vmax you have to use other register, E4 0 before and E4 to 1 after the 3D register.
@MixedC confirmed here that is working with vrmtool: https://www.overclock.net/forum/27096897-post612.html so it could work with afterburner.


Edited:
I found a higher resolucion image of you pcd: https://www.ixbt.com/video3/images/a...scan-front.jpg

In my attached image you can see the two vrms for the vddci and the memory, I see two littel chips highlighted in green that could be controlling them, so maybe the hwinfo vddci is only one of the core phases. But I can not confirm this because I the name of the chips can't be seen.

Maybe somebody here could hepl you to know if these chips are controlling the vddci and memory vrms, I can not do more. The other way to know the chips name is to open your gpu and see it.
I tried writing to the registers with MSI Afterburner and it works as expected.



+75mV VDDC and +50mV VDDCI gets limited to 1.244v VDDC and 1.138v VDDCI as set by Vmax.

I tried adding the registers to my VOI following gupsterg's guide on adding VDDC offset to the BIOS, but I get a black screen after I restart.

I checked the data tables and they've all moved by 8 bytes as expected (I changed register 0x8D to 0x3D in the VOI and added 2 0xE4 registers), so I'm not sure what's wrong.

I attached my BIOS so if you can figure out what I did wrong, that would be kindly appreciated.
380_modded_2_1500+1750-stilt-afr_Vmax is the BIOS with the Vmax and 0xE4 registers added.
380_modded_2_1500+1750-stilt-afr is my current BIOS.

VOI:
Code:
6A 00 03 01 01 07 0C 00 0E 00 00 00 00 00 00 00 
04 00 24 00 00 04 00 00 00 80 10 00 00 00 00 00 
39 03 00 00 10 00 B6 03 00 80 00 00 CF 03 00 80 
10 00 4C 04 05 03 22 00 08 96 60 00 00 00 00 00 
14 00 2A 00 E4 00 00 00 3D 00 43 00 E4 00 01 00
8E 00 00 00 22 00 78 00 23 00 78 00 FF 00 05 07 
0C 00 00 00 00 00 00 00 00 00
Spoiler!


The top VRM controller is UP1540P. Not sure on the bottom.

Thanks so much.
Attached Files
File Type: zip 380_modded_2_1500+1750-stilt-afr_Vmax.zip (100.2 KB, 2 views)
File Type: zip 380_modded_2_1500+1750-stilt-afr.zip (100.2 KB, 2 views)


Last edited by BLUuuE; 12-28-2018 at 04:42 AM.
BLUuuE is offline  
post #1107 of 1125 (permalink) Old 12-28-2018, 08:28 AM
New to Overclock.net
 
mynm's Avatar
 
Join Date: Jan 2016
Posts: 623
Rep: 54 (Unique: 20)
You are welcome

Thank for the image and the info.

Knowing that is a uP1540P, I have just found this review http://greentechreviews.ru/2015/10/2...-strix-gaming/

In this image can be seen clearly that both VRMs are using the uP1540P http://greentechreviews.ru/wp-conten...015/10/218.jpg

And acording to the traslator it says that: "The memory power subsystem has two phases (each has its own driver)." "And the uP1540P controller is responsible for the voltage". I suppose that one is for the vdcci and the other for the memory and not two phases for the memory.

So hwinfo vddci is not the vddci, is only one phase for the core, but I have no idea why is used like this.

Maybe is used to reduce the OCing capability with software, because the first loop voltage offset is not affecting the second loop voltage with software, but it is affected setting an offset voltage into the VOI, if I remember correctly.

I have no idea if uP1540P could be controlled by afterburner.

About the VOI, you forget to change the length to 2A, I don't see more errors.

6A 00 03 01 01 07 0C 00 0E 00 00 00 00 00 00 00
04 00 24 00 00 04 00 00 00 80 10 00 00 00 00 00
39 03 00 00 10 00 B6 03 00 80 00 00 CF 03 00 80
10 00 4C 04 05 03 2A 00 08 96 60 00 00 00 00 00
14 00 2A 00 E4 00 00 00 3D 00 43 00 E4 00 01 00
8E 00 00 00 22 00 78 00 23 00 78 00 FF 00 05 07
0C 00 00 00 00 00 00 00 00 00

Edited: Maybe with a I2C dump you could see the I2C tables for the uP1540P controllers, and maybe figuring out the voltages if the I2C table is similar to other uP controllers of the afterrburnet Third party Graphics Card hardware database. Like here: https://forums.guru3d.com/threads/th...3#post-5091528

Last edited by mynm; 12-28-2018 at 08:48 AM.
mynm is offline  
post #1108 of 1125 (permalink) Old 12-28-2018, 04:03 PM
@integralfx
 
BLUuuE's Avatar
 
Join Date: Feb 2014
Posts: 195
Rep: 7 (Unique: 6)
Quote: Originally Posted by mynm View Post
You are welcome

Thank for the image and the info.

Knowing that is a uP1540P, I have just found this review http://greentechreviews.ru/2015/10/2...-strix-gaming/

In this image can be seen clearly that both VRMs are using the uP1540P http://greentechreviews.ru/wp-conten...015/10/218.jpg

And acording to the traslator it says that: "The memory power subsystem has two phases (each has its own driver)." "And the uP1540P controller is responsible for the voltage". I suppose that one is for the vdcci and the other for the memory and not two phases for the memory.

So hwinfo vddci is not the vddci, is only one phase for the core, but I have no idea why is used like this.

Maybe is used to reduce the OCing capability with software, because the first loop voltage offset is not affecting the second loop voltage with software, but it is affected setting an offset voltage into the VOI, if I remember correctly.

I have no idea if uP1540P could be controlled by afterburner.

About the VOI, you forget to change the length to 2A, I don't see more errors.

6A 00 03 01 01 07 0C 00 0E 00 00 00 00 00 00 00
04 00 24 00 00 04 00 00 00 80 10 00 00 00 00 00
39 03 00 00 10 00 B6 03 00 80 00 00 CF 03 00 80
10 00 4C 04 05 03 2A 00 08 96 60 00 00 00 00 00
14 00 2A 00 E4 00 00 00 3D 00 43 00 E4 00 01 00
8E 00 00 00 22 00 78 00 23 00 78 00 FF 00 05 07
0C 00 00 00 00 00 00 00 00 00

Edited: Maybe with a I2C dump you could see the I2C tables for the uP1540P controllers, and maybe figuring out the voltages if the I2C table is similar to other uP controllers of the afterrburnet Third party Graphics Card hardware database. Like here: https://forums.guru3d.com/threads/th...3#post-5091528
I updated the length to 0x2A in the VOI and it works perfectly now.

I attached my I2C dump but there doesn't seem to be any UP1540P devices.

Code:
------[ Chips Found ]------

GPU1-B04-D3A: ***** Unknown Device *****
GPU1-B04-D50: EDID
GPU1-B06-D30: IRF IR3567B
GPU1-B06-D70: IRF IR3567B
GPU1-B06-D71: IRF IR3567B
Attached Files
File Type: txt atismbusdump.txt (19.0 KB, 2 views)

BLUuuE is offline  
post #1109 of 1125 (permalink) Old 12-28-2018, 05:12 PM
New to Overclock.net
 
mynm's Avatar
 
Join Date: Jan 2016
Posts: 623
Rep: 54 (Unique: 20)
Quote: Originally Posted by BLUuuE View Post
I updated the length to 0x2A in the VOI and it works perfectly now.

I attached my I2C dump but there doesn't seem to be any UP1540P devices.

Code:
------[ Chips Found ]------

GPU1-B04-D3A: ***** Unknown Device *****
GPU1-B04-D50: EDID
GPU1-B06-D30: IRF IR3567B
GPU1-B06-D70: IRF IR3567B
GPU1-B06-D71: IRF IR3567B

Is great to see the Vmax mod working.

I can't understand the dump, so it have three I2C tables for the IR3567B, the one you are using is at 6:30h like here:

https://forums.guru3d.com/threads/th...7#post-5222098

But that dump don't have the other two tables.

Maybe the 70 and 71 ones are the UP1540P ones, as they don't look like a IR3567B I2C table, but they also don't look like the uP6266 one, and Afterburner does not support the UP1540P. So I don't know how to see the the voltages if UP1540P is a voltage controllers with LUT registers.

So we only can try to guess how to see the voltages, and try to find info. But maybe I am wrong and are not UP1540P I2C tables. This is to complicated for me.
mynm is offline  
post #1110 of 1125 (permalink) Old 12-28-2018, 11:10 PM
@integralfx
 
BLUuuE's Avatar
 
Join Date: Feb 2014
Posts: 195
Rep: 7 (Unique: 6)
Quote: Originally Posted by mynm View Post
Is great to see the Vmax mod working.

I can't understand the dump, so it have three I2C tables for the IR3567B, the one you are using is at 6:30h like here:

https://forums.guru3d.com/threads/th...7#post-5222098

But that dump don't have the other two tables.

Maybe the 70 and 71 ones are the UP1540P ones, as they don't look like a IR3567B I2C table, but they also don't look like the uP6266 one, and Afterburner does not support the UP1540P. So I don't know how to see the the voltages if UP1540P is a voltage controllers with LUT registers.

So we only can try to guess how to see the voltages, and try to find info. But maybe I am wrong and are not UP1540P I2C tables. This is to complicated for me.
Don't worry about it then.



I seem to be getting a few memory errors, but I was mostly browsing the internet and using Discord. Nothing too intensive.

I think it might be the low VDDCI, but I'm not too sure how I would change it.

I tried setting usVddcOffset for DPM0-6 but it doesn't change the VDDCI when idling. It's still 975mV, which is strange since changing DPM7 usVddcOffset changes the VDDCI under load.

If what you say about VDDCI actually not being the real VDDCI is true, then would that mean the real VDDCI is locked and can't be changed from BIOS? That seems to be the case here.

I found out that my BIOS is old.

I ended up flashing this BIOS and am now testing it. If all my issues before was because of an old BIOS... (╯°□°)╯︵ ┻━┻


Last edited by BLUuuE; 12-29-2018 at 12:34 AM.
BLUuuE is offline  
Reply

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off