Hawaii Bios Editing ( 290 / 290X / 295X2 / 390 / 390X ) - Page 85 - Overclock.net - An Overclocking Community

Forum Jump: 

Hawaii Bios Editing ( 290 / 290X / 295X2 / 390 / 390X )

Reply
 
Thread Tools
post #841 of 4258 (permalink) Old 01-05-2016, 01:39 PM
New to Overclock.net
 
The Stilt's Avatar
 
Join Date: May 2013
Posts: 2,270
Rep: 750 (Unique: 286)
Quote:
Originally Posted by Stige View Post



plz tell me how I old it wronk frown.gif so I can hold it right, and overclock it to the .. something!

The display driver can completely override your bios PowerPlay table, if certain tables exist. Uninstall all the crap (Afterburner and all similar software) and clean the display driver with DDU. The flash in one of the bios versions I sent you and reinstall the drivers. If it still doesn´t work, then try following:

- Flash some bios version (ATIFlash)
- Dump the bios in Windows using ATIWinFlash or GPU-Z
- Compare the checksums of the input (flashed) and the readout file

If the checksums don´t match, you´ll know where the problem is. If they do match and you´ve removed all the 3rd party software (Afterburner, etc) and the display drivers (fully), then there might be a configuration issue with your card wink.gif
The Stilt is offline  
Sponsored Links
Advertisement
 
post #842 of 4258 (permalink) Old 01-05-2016, 01:54 PM
New to Overclock.net
 
The Stilt's Avatar
 
Join Date: May 2013
Posts: 2,270
Rep: 750 (Unique: 286)
Quote:
Originally Posted by OneB1t View Post

implemented memory timing editor
it can have some problems so use with caution (recheck by reloading into hawaiibiosreader again to control values or by hand in hex)




if someone can help me with identifiing each field it will be nice (ras, cas, etc..)
maybe this can help little (but maybe its not for GDDR5)

Those timings are certainly not for GDDR5. They appear to be the same as the ones available in RaBit (DDR). None of the newer ATOM related tools support adjusting memory timings, since the memory blocks (profiles for each ASIC and memory type) are selected from the GoldenDB (AMD server). Without the register descriptions (for each ASIC) and ability to monitor EDC, blindly tampering with the timings do more harm than good.
The Stilt is offline  
post #843 of 4258 (permalink) Old 01-05-2016, 01:57 PM
New to Overclock.net
 
Join Date: Jun 2011
Posts: 710
Rep: 49 (Unique: 30)
is it dangerous somehow or just can lead to instability?

about EDC errors i can prolly write some quick&dirty openCL kernel to load GPU memory and intergrated ddr driver to 100%
then check performance if its lower then EDC errors are slowing down whole card

OneB1t is offline  
Sponsored Links
Advertisement
 
post #844 of 4258 (permalink) Old 01-05-2016, 02:09 PM
New to Overclock.net
 
The Stilt's Avatar
 
Join Date: May 2013
Posts: 2,270
Rep: 750 (Unique: 286)
Quote:
Originally Posted by navjack27 View Post

since you are posting so much i'd really like to ask you a question then. i did a crappy little video on frame timings. i used different memory timings and overclocks to see what did what to csgo. i consider the game to be very well optimized and i play it a bunch so i like using something i'm familiar with. i noticed that when i went from stock timings with the same core and memory clocks and changed the timings by taking 1250 timings and moving them to every strap higher that i got tighter frame times and less stuttering.

would corrected errors show up in a test like this? why would tighter timings help in frame latency? why do i see a strap for 2000mhz on the memory in my bios? i guess i'd also like to learn more about powerplay if you know anything about that... does disabling it in afterburner void out any power limit settings?

I know nothing about gaming, since I personally play games maybe once or twice per year biggrin.gif

While keeping that in mind:

Have you verified the results by using different, verified (EDC free) timings at the same memory clock?

Use stock, native bios for your card and make the measurements. Then edit the bios in a way that you copy the timings for the next frequency range to previous one (for the frequency which you used in the tests). Does the higher latency make a difference which can be constantly reproduced?

E.G.

Original

1250 = DEADDEAD
1375 = EFBEEFBE

Test

1250 = EFBEEFBE
1375 = EFBEEFBE
The Stilt is offline  
post #845 of 4258 (permalink) Old 01-05-2016, 02:11 PM
New to Overclock.net
 
The Stilt's Avatar
 
Join Date: May 2013
Posts: 2,270
Rep: 750 (Unique: 286)
Quote:
Originally Posted by OneB1t View Post

is it dangerous somehow or just can lead to instability?

It might render the card unbootable, but nothing more. At least I´ve never managed to kill anything.
The Stilt is offline  
post #846 of 4258 (permalink) Old 01-05-2016, 02:12 PM
New to Overclock.net
 
Join Date: Jun 2011
Posts: 710
Rep: 49 (Unique: 30)
thats what i wanted to hear
this is not really about performance anymore.. its just about playing with toys and using their potencial to limits
you can get whole week memory performance tweaks by increasing gpu clock by 10Mhz biggrin.gif

OneB1t is offline  
post #847 of 4258 (permalink) Old 01-05-2016, 02:21 PM
New to Overclock.net
 
jdorje's Avatar
 
Join Date: Oct 2014
Posts: 1,351
Rep: 39 (Unique: 32)
Since I've been playing around my valley scores are about 50 points lower. I've tried reverting drivers, going back to beta drivers, disabling igpu, changing bios back to original and back, closing everything in the background...my stock scores won't go over 2600 (maxed at 2848 before) and my overclock scores are all in line with that.

Kombustor might also be affected, but firestrike and heaven are not.

Haven't tried uninstalling afterburner, I guess. I've deleted the profiles many times however.

What gives?

jdorje is offline  
post #848 of 4258 (permalink) Old 01-05-2016, 02:22 PM
New to Overclock.net
 
Join Date: Jun 2011
Posts: 710
Rep: 49 (Unique: 30)
try DDU

OneB1t is offline  
post #849 of 4258 (permalink) Old 01-05-2016, 02:29 PM
New to Overclock.net
 
jdorje's Avatar
 
Join Date: Oct 2014
Posts: 1,351
Rep: 39 (Unique: 32)
Quote:
Originally Posted by OneB1t View Post

try DDU

Ah yes, I left off the list: I ddu uninstalled and reinstalled drivers then did the same for beta drivers. No change.

I did notice if you run valley windowed it gives lower fps. But if you run it full screen and tab out and back in it also gives that same lower fps. That's about a 50 point difference (2%) also.

jdorje is offline  
post #850 of 4258 (permalink) Old 01-05-2016, 02:32 PM
New to Overclock.net
 
Join Date: Jun 2011
Posts: 710
Rep: 49 (Unique: 30)
maybe cpu overclock? that also have some impact on unigine scores (and pci-e version...)

OneB1t 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