Overclock.net banner
Cancel
Create thread New Forums

A4000 shunt mod. Is it possible? Can we get 3070Ti performance in 3060Ti Aero footprint?

Jump to Latest
21 - 34 of 34 Posts
#21 ·
menthu_rae said:
Hi all - I joined up to post in here - I am a fellow "A4000 + Palit heatsink + 8-pin soldered" convert. I was previously running the MSI 3060 Ti Aero ITX OC but found the 8GB VRAM limiting for AI applications and also I would have to throttle the power limit down to avoid the fan getting too loud/it running too hot.

I just wanted to say fantastic work to all involved with the shunt mod! I had been wondering if it was possible after seeing it done on the A2000 with good results. Glad to see that it is. :)

I'm also looking for a bit of assistance. I have found the card I've bought, while it runs "fine", it appears to be downclocking itself. Even when running <50°C (122°F) I am getting core / mem of 1425 / 6500 - which seems low.

I have manually tweaked the card in afterburner to +200/500 core/mem and that gives me ~1500-1600 core (fluctuates) and a solid 7000 memory.

Does anyone know if it's possible the BIOS was modded on this (likely ex-mining) card? Does anyone have a guaranteed stock BIOS I could try flashing on?

Thanks for any help anyone can provide. This card is so close to perfect for everything I currently do and I'm glad to see others enjoying this little beast!
Click to expand...
These BIOS can't be modified, I highly doubt you're seeing a BIOS issue here. This sounds more like the BIOS balancing performance under the stock power limit constraints.

I have 4 different BIOS for these, one from each vendor of the card. You are welcome to try them if you'd like, contact us through our website and I'll send them over: A4000 Power Mod with SFF Cooler - JK&G Ventures
 
#22 ·
jkgventures said:
Well guys, we upped our score to 14123 which soundly matches a stock 3070ti.

we have officially listed this as a service we offer on our website now, extensive testing has shown these will work well in the long term with the SFF cooler and the custom brackets we developed to provide adequate clamping pressure for the VRM thermal pads.

A4000 Power Mod with SFF Cooler – JK&G Ventures

jkgventures.tech jkgventures.tech
Click to expand...
Awesome news. Are you willing to share some more information at all on the VRM clamping bracket? Or even perhaps sell this as a standalone part?

jkgventures said:
These BIOS can't be modified, I highly doubt you're seeing a BIOS issue here. This sounds more like the BIOS balancing performance under the stock power limit constraints.

I have 4 different BIOS for these, one from each vendor of the card. You are welcome to try them if you'd like, contact us through our website and I'll send them over: A4000 Power Mod with SFF Cooler - JK&G Ventures
Click to expand...
OK that's good to know - and yes I think you are right. I'll shoot you a message and take a look at the BIOS checksums against the one on my card anyway. Thank you! (y)
 
#24 ·
Hey everyone 👋

I've made a video including all of my test results after sending my card into JK&G Ventures. Check it out, it's pretty crazy to see the performance increase on the RTX A4000. It truly is a mini RTX 3070 Ti!

Also, John from JK&G Ventures was kind enough to provide a coupon code for the viewers of this video to get 15% off your order of the A4000 Power Mod. The code is in the video description! So, if you're interested in the mod, why not save a little bit of money on it 😉

 
#25 ·
NattyKathy said:
Absolutely possible. Be careful with the VRM, I probably wouldn't go over 900mV-950mV sustained VCore under load if I were doing this. At various points last year I was strongly considering getting an A4000 and modding it but ended up not- I have some thoughts though.

(teardown photos source- Reddit)
(markup by me)
View attachment 2595837

Like I said, mind the VRM if you power mod- make sure Power Stages and inductors are well-cooled and keep an eye on the GPU Core-only power usage (NVDD). It looks like the actual NVDD VRM section is only 5 phases (note that WCCFTech cited it as 6-phase which appears to be incorrect) and assuming this is using 50A 55A Power Stages like the A2000, 3070 FE, 3070Ti FE, and others (AOZ5311NQI or similar), that will be a significant factor in limiting max sustainable clocks. Here's a snippet of the efficiency curves for a AOZ5311NQI with markups for NVDD output based on 5 phases and 1V VCore. Remember that Power Stage ratings are absolute maximums! The most comfortable zone for a VRM is around 1/3 of max capacity but running at 1/2 to 2/3 capacity can be okay if VRM cooling is sufficient- note the exponential curve in heat output as switched current rises and efficiency drops.

NVDD Output Wattage vs VRM efficiency- 5 Phases x 50A 55A
View attachment 2595842

The "good" news (for safety) is that stacking 8mΩ shunts on top of the existing 5mΩ should result in a new Board Power Limit of only around 230W which ought to be safe-ish for the VRM as long as it's well-cooled (since NVDD PL is lower than the Board Power) while still giving a nice boost.
Stacking 5mΩ should result in approx 280W PL which is either unsafe or catastrophic for that VRM depending what share of the Board Power goes to NVDD vs GDDR but could be useful if you want to "float" the V/F curve and manage throttling and PL yourself while still having somewhat of an upper limit.
2mΩ stacked would be unwise as that's like 500W- effectively "completely uncapped" territory for GA104, you would have to have a carefully managed V/F curve set at all times as that mod would put the card in a state where it could now very easily damage/destroy a power stage at stock with sudden load applied.

Anyways, hope all this rambling is informative. I think the A4000 is an interesting card and I still vaguely would like to mod one myself someday, maybe when they're available used for cheap in a couple years. Please do post updates if you decide to power mod yours and be nice to the VRM, it will be working quite hard!

edit to add- another thought comes to mind- what is the default V/F curve for that card like? It occurs to me that it may not even be able to call for higher voltage, can you check that? I'd assume it can, but the thing was clearly intended to run at extremely low VCore...
Click to expand...
What about stacking R015 on PCIE and one more R005 on 12V? this would make 100W from PCI-E slot and 130W from cable? Would it be safe, or it's too much to draw from cable since there is only 1 connector? There is no need to stack on NVDD?
 
#26 ·
Buko84 said:
What about stacking R015 on PCIE and one more R005 on 12V? this would make 100W from PCI-E slot and 130W from cable? Would it be safe, or it's too much to draw from cable since there is only 1 connector? There is no need to stack on NVDD?
Click to expand...
I do not think that method would succeed to use more power from the slot. The shunt resistors are only there to help enforce a maximum cap on current draw from each separate input rail- the actual amount of current pulled from each input at any time is balanced by the BIOS & driver. Recent NV cards (Turing and Ampere at least based on firsthand experience) draw very little power from the PCIe slot regardless of shunt mods.
On cards that have any auxiliary 12V input, the PCIe slot 12V rail is sometimes not even connected to the GPU Core VRMs at all- instead exclusively powering other rails such as GDDR VRMs and/or minor rails such as 5V / 1.8V etc while the aux 12V input(s) power the Core alone. On some cards the PCIe slot 12V rail is connected to some of the Core VRM but the BIOS / Driver will choose to only use a little of that power, prioritizing using power from the auxiliary 12V rails.

TL;DR power rail balancing on NV cards is complex and it is best to only consider the auxiliary 12V inputs when estimating power targets. Shunt modding the PCIe slot pwr rail is only done to keep offsets "in balance" and shouldn't really increase reliance on slot power.
 
  • Helpful
Reactions: Paradigm Shifter
Save
Reply Quote Rep+ Rep+
#27 ·
NattyKathy said:
The "good" news (for safety) is that stacking 8mΩ shunts on top of the existing 5mΩ should result in a new Board Power Limit of only around 230W which ought to be safe-ish for the VRM as long as it's well-cooled (since NVDD PL is lower than the Board Power) while still giving a nice boost.
Stacking 5mΩ should result in approx 280W PL which is either unsafe or catastrophic for that VRM depending what share of the Board Power goes to NVDD vs GDDR but could be useful if you want to "float" the V/F curve and manage throttling and PL yourself while still having somewhat of an upper limit.
2mΩ stacked would be unwise as that's like 500W- effectively "completely uncapped" territory for GA104, you would have to have a carefully managed V/F curve set at all times as that mod would put the card in a state where it could now very easily damage/destroy a power stage at stock with sudden load applied.
Click to expand...
i just wanted to take more power from 12V since that shunts won't result more than 230W. Will the card not pull more from the plug or will it be dangerous for it? Stacking 8mΩ on PCIe will result about 122W while around 106W from plug. I thought I could increase the consumption from the cable by giving a smaller resistor to the pcie socket but not to exceed 230W overall.
 
#28 ·
NattyKathy said:
Shunt modding the PCIe slot pwr rail is only done to keep offsets "in balance" and shouldn't really increase reliance on slot power.
Click to expand...
so we should shunt mod that slot or not if it does nothing? what would happen if card would take that 12x W of power from slot? or maybe that 75W is slot limit? I am confused, that card above has a shunt also on PCIe
 
#30 ·
Buko84 said:
so we should shunt mod that slot or not if it does nothing? what would happen if card would take that 12x W of power from slot? or maybe that 75W is slot limit? I am confused, that card above has a shunt also on PCIe
Click to expand...
I'll try explaining a little differently if my previous post was confusing.

Recent NVidia GPUs will not pull lots of power from the PCIe slot if they have 12V plugs, this is a choice in GPU BIOS and often reflected in the hardware design where NVDD VRMs may not even be connected to the Slot power rail. If you need more power, draw it from the 12V plug. In this case, if the 230W limit from stacked 8mΩ is not enough, the choice is to use stacked 5mΩ instead for 280W limit, make sure the VRM is well-cooled, and watch your voltages.

But YES I think modding the slot regardless is a good idea.

The opinion of many people who have Shunt Modded Nvidia Turing / Ampere cards is that modding the slot shunt is a good idea even if it does not directly increase slot power consumption.

This is because the GPU expects a certain ratio of power draw from each power source. If the 12V plug(s) are shunt modded but not the slot, the GPU may think that the ratio is off and this can cause trouble.

Regarding slot capacity, 75W is the rating of the PCIe slot specification per PCI-SIG standards. This is not a software / hardware limit set by the motherboard but a safety rating- it is up to the GPU to respect that limit. Some do not- for example back in 2016 there was trouble with some Radeon 480 pulling too much power from the slot and damaging motherboards. Around 100W from the slot is usually ok but this depends on the motherboard. Not every motherboard can handle high power draw from a single PCIe slot!
 
  • Helpful
Reactions: Paradigm Shifter
Save
Reply Quote Rep+ Rep+
#31 ·
I modified my A4000 by adding 8mOhm on pcie slot and 8 pin connector, but how can I check if it was done correctly? I forgot to do tests before shunt and cooling mod :/ in hwinfo I don't even see bigger results than ~150W. Maybe my soldering skills need to be corrected a bit, or I also need to put a shunt on NVDD to make it work?

Photograph Product Font Screenshot Software
 
#32 ·
Hi all. I did a little research on the subject and found out that my initial shunt was made a little wrong and the conclusions that I made at the beginning are not entirely correct.
What we have: 3 resistor shunts. I left only 2. On pci and 8pin. The shunt is working, the indicators are 110x1.62=180w. The total power of the card is limited to 110w in osd.
I found a shunt mod calculator on github and after studying it a little, I concluded that despite the fact that we are shunting the pci slot, its consumption does not increase and therefore the power is limited to 110w. Thus, we get an honest consumption of 180w as indicated in the shunt calculator without using a pci shunt. Thus, changing the resistors to a lower value of 6mOm will give a gain of up to 194w. I checked with myself and the frequencies grew even higher, but then I still changed it back to 8mOm for safety. I also have limited msi afterburner on 1830mhz/825mv/~153w
To get the correct multiplier parameters, you need to check the pci shunt checkbox, for the rest of the parameters, the checkbox must be unchecked

github.com

GitHub - bmgjet/ShutMod-Calculator: Work out what shunt values to use easily.

Work out what shunt values to use easily. Contribute to bmgjet/ShutMod-Calculator development by creating an account on GitHub.
github.com github.com
 

Attachments

#34 ·
Buko84 said:
Why did you changed it back? Was it going too hot? I am thinking about doing 7mOhms, maybe it won't be so dangerous for the card.
Click to expand...
I have a small case. With a long load, I can not measure the temperature in it. In testing, temperatures were not too high at 6mOm. Just in case, I left 8mOm What would be colder. I don't really need to overclock the card much. What I do is kind of a hobby) I have ~75c with 160w in small case. in open stand -6c. i also meseared vrm on 1 minutes furmark on back side of gpu and it was near 45c on 6mOM.
 
21 - 34 of 34 Posts
About this Discussion
33 Replies
10 Participants
Last post:
sanakess
Overclock.net
posts
28M
members
562K
Since
2004
The premier forum for overclocking experts and enthusiasts. Discuss hardware optimization, custom builds, benchmarking, cooling solutions, and pushing the boundaries of computing performance. From beginner guides to extreme overclocking, join our technical community to master system tuning.
Full Forum Listing
Explore Our Forums
Intel CPUs Hardware News NVIDIA PC Gaming Water Cooling