Question GIGABYTE 5700 XT Bios mod fails

AlleyCat

Veteran
Mitglied seit
Okt 9, 2020
Beiträge
290
Bewertungspunkte
17
Punkte
17
Hi,
I am failing to update Gigabyte 5700 XT. The same procedure updates the BIOS on MSI cards with no problem. I follow the instructions on Igors Lab.

The sign of having trouble with the card bios flash shows in GPU-Z. After flash using amdvbflash the values of GPU and memory frequencies is empty. With stock bios there are Mhz frequencies.

Is there any known problems with flashing Gigabyte cards?
Any suggestions on what other forums I may ask for assistance?
Is it possible that the OEM bios is signed, and any modified bios will be rejected?
If the bios is signed, any tools to resign, or would I need to buy a card from a different vendor?

Thanks,

Alley Cat
 
Zuletzt bearbeitet :
no is not work 1400
only work in 1390 900
is the reference card with blower Samsung memory
1350 890 54.7m
1350 900 55.3
but Unstable
This is my configuration for the AMD Radeon Reference 5700XT Samsung memory.
The speed is 55.60 with Red Team Miner. Phoenix miner will report a higher hash rate.
The card is reasonably stable. It crashed only once in 24 hours with 830mv, so I increased to 840mv, and now testing for stability. .
 

Anhänge

  • Screen Shot 2020-11-09 at 7.25.23 PM.png
    Screen Shot 2020-11-09 at 7.25.23 PM.png
    56 KB · Aufrufe : 61
This is my configuration for the AMD Radeon Reference 5700XT Samsung memory.
The speed is 55.60 with Red Team Miner. Phoenix miner will report a higher hash rate.
The card is reasonably stable. It crashed only once in 24 hours with 830mv, so I increased to 840mv, and now testing for stability. .
bios What is used?
I try it thx
 
This is my configuration for the AMD Radeon Reference 5700XT Samsung memory.
The speed is 55.60 with Red Team Miner. Phoenix miner will report a higher hash rate.
The card is reasonably stable. It crashed only once in 24 hours with 830mv, so I increased to 840mv, and now testing for stability. .
i test it
one minutes
a card crashed
 
amd the reference card with blower Samsung memory
so is work
1350-1395
maybe 900 work in 890-875
I didn't do mpt
in trm is work 1350 890 725 700 54.3
but is Unstable
 
This is my configuration for the AMD Radeon Reference 5700XT Samsung memory.
The speed is 55.60 with Red Team Miner. Phoenix miner will report a higher hash rate.
The card is reasonably stable. It crashed only once in 24 hours with 830mv, so I increased to 840mv, and now testing for stability. .
AlleyCat, You could stick with this power table and vram timings below,

[Updated]
This is the final power table for any RX 5700 card in the kindly attached link below,

 
Zuletzt bearbeitet :
O
AlleyCat, stick with this power table and vram timings and change the memory timing controller to 1 for cards with Samsung memory as below,

- Under Overdrive Limits tab, set:
GFX Maximum Clock 1500 MHz
Memory Maximum Clock 1000 MHz
Power Limit Maximum 0%
Power Limit Minimum 0%
Memory Timing Control 1 [Recommended for cards with Samsung memory]

- Under Power and Voltages, set:
Maximum Voltage GFX 850 mV
Maximum Voltage SoC 1150 mV
Minimum Voltage GFX 700 mV
Minimum Voltage SoC 700 mV
Power Limit GPU 140 W
TDC Limit GFX 150 A
TDC Limit SoC 12 A

- Under Frequency tap, set:
GFX Maximum 1150 MHz
Memory DPM 3 - 950 MHz

V. Under Fan tab, set:
Stop Temperature 60° C
Start Temperature 70° C

- K4Z80325BC (Samsung)
(1500 or 1550) MHz (Copy) - - - > (Paste) 1800 MHz
1800 MHz - - - > 2000 MHz
2000 MHz - - - > 2250 MHz

Memory Clock - tREF
1000 MHz - 3900
1250 MHz - 4875
1550 MHz - 6045
1800 MHz - 7020
2000 MHz - 7800
2250 MHz - 8775

and test the memory clock at 930 MHz, VDDCI at 900 mV and MVDD at 1320 mV.
Ok. Will do it tomorrow.
 
This is a nice thread helping everyone squeeze the max out of our cards. I'd just like to take a second and give one piece of advice based on my experience:
1) get your rig STABLE so you can forget about it and just let it make coins. This extends to pool choice and actual share number as well, not only miner reported rate and stability.
2) DON'T mess with it once it's working. Any little thing can throw it off. Hooking up mice, keyboards, monitors...I now went to test one card and after nice 10 hours of solid work, after punching in the DEFAULT voltage number for one card, another card fell off and the rig restarted. Crazy.
 
I tried modifying 5700 bios on a 5700 XT but saw no improvement in performance. In my opinion, keeping consistency with modified original bios is preferred for television management, unless I find a noticeable change in performance.

How significant on your system is the performance difference between an XT and non-XT bios?

@Mini_Me, did you find what may be the differences in the bios?

Thanks,
AlleyCat
about 10w less on the software for each gpu and stability also seems better with the same configuration.

1604988242723.png
 

Anhänge

  • rig-0-RadeonRX5700-8G-Micron_GDDR6-113-3E4112U-S48.rom
    1 MB · Aufrufe : 17
This is a nice thread helping everyone squeeze the max out of our cards. I'd just like to take a second and give one piece of advice based on my experience:
1) get your rig STABLE so you can forget about it and just let it make coins. This extends to pool choice and actual share number as well, not only miner reported rate and stability.
2) DON'T mess with it once it's working. Any little thing can throw it off. Hooking up mice, keyboards, monitors...I now went to test one card and after nice 10 hours of solid work, after punching in the DEFAULT voltage number for one card, another card fell off and the rig restarted. Crazy.

Here what I do to stabilize the rig :
1) Update PowerTables and memory straps.
2) Test each new card in a test rig. This provided me with a range of performance for different brands. Since I am having multiple types of cards, I tested each brand or type. The same brand may have a different memory, and its range will be specific for a memory type.
3) Add cards to the rig, one at a time. I attempt to use the same OC settings for similar cards.
4) I create an OC profile as a template and give it a revision number. For example, PRODUCTION-8.1. Eight cards, rev. 1.
5) When the miner reports a crash, I modify only that card and save a new template as PRODUCTION-8.2. Eight cards, rev. 2.
6) I use a miner with default settings. All my cards are AMD, and the Red Team Miner is my preferred miner. The RTM reports accurate hash and power, and I prefer real numbers for calculating the rig's efficiency over a long period.
7) Crashed cards require higher VDD. Invalid shared needs less memory OC, or move VDDCI.
 
这是我为稳定钻机所做的工作:
1)更新PowerTables和内存条。
2)在测试装置中测试每个新卡。这为我提供了不同品牌的一系列表现。由于我有多种类型的卡,因此我测试了每个品牌或类型。同一品牌可能具有不同的内存,并且其范围将特定于内存类型。
3)一次向卡中添加卡片。我尝试对相似的卡使用相同的超频设置。
4)我创建一个OC配置文件作为模板,并给它一个修订号。例如,PRODUCTION-8.1。八卡,转。1。
5)当矿工报告崩溃时,我仅修改该卡并将新模板另存为PRODUCTION-8.2。八卡,转。2。
6)我使用默认设置的矿工。我所有的卡都是AMD,红队矿工是我的首选矿工。RTM报告准确的哈希值和功率,我更喜欢使用实数来计算钻机的长期效率。
7)损坏的卡需要更高的VDD。无效的共享需要较少的内存OC或移动VDDCI。
我发现了一些有趣的现象
现代BIOS
所有稳定性都指向合理的电源值
很难找到一个合理的价值,有时这只是一个关键点
Sometimes this is futile and stability is greater than the cost of some electricity
 
Here what I do to stabilize the rig :
1) Update PowerTables and memory straps.
2) Test each new card in a test rig. This provided me with a range of performance for different brands. Since I am having multiple types of cards, I tested each brand or type. The same brand may have a different memory, and its range will be specific for a memory type.
3) Add cards to the rig, one at a time. I attempt to use the same OC settings for similar cards.
4) I create an OC profile as a template and give it a revision number. For example, PRODUCTION-8.1. Eight cards, rev. 1.
5) When the miner reports a crash, I modify only that card and save a new template as PRODUCTION-8.2. Eight cards, rev. 2.
6) I use a miner with default settings. All my cards are AMD, and the Red Team Miner is my preferred miner. The RTM reports accurate hash and power, and I prefer real numbers for calculating the rig's efficiency over a long period.
7) Crashed cards require higher VDD. Invalid shared needs less memory OC, or move VDDCI.
I found some interesting phenomena
Mod BIOS
All stability points to a reasonable power supply value
It is difficult to find a reasonable value, sometimes this is just a key point
Sometimes this is futile and stability is higher than the cost of some electricity
 
Oben Unten