win32asmguy
-
Posts
531 -
Joined
-
Last visited
-
Days Won
3
Content Type
Profiles
Forums
Events
Everything posted by win32asmguy
-
I think it is power limits supplied by Dell within Nvidia's range for the product. The driver is supposed to enforce those limits via the PCF component. I think this driver also has other bugs (noticed some artifacting even at stock clocks) so I reverted back to the Dell provided drivers for now as I do not want to damage the GPU.
-
Nice! I guess the next question is if Nvidia will be patching this work around in the next driver release. It sounds like it may be similar to what MSI does with Overboost and Tongfang with their TDP unlocks. It may also be something related to Nvidia allowing higher TDP on the desktop 4090 card too.
-
Yes, in Timespy it maxes out at 125W. In games I regularly see 120W+ now where before it was only 102W with the Dell provided Nvidia PCF driver.
-
Tonight I managed to get the 3080Ti in the 7670 working at 125W by using MSI Afterburner to control power limits intead of Nvidia PCF. Here is a 3dmark score: https://www.3dmark.com/3dm/86161666? Basic process, I think: Be on bios 1.8.0, with Nvidia driver 527.56, msi afterburner 4.6.5 beta 2. In Device Manager, disable "Software Devices -> NVIDIA Platform Controllers and Framework". Open MSI Afterburner. The power limit shown will likely be a nonsense value, drag the slider to 1% then back up to 138%, then Apply the changes. Now the higher GPU power draw over 102W should be observed.
-
The Zbook Fury 16 G9 is 90W for the A4500, no dynamic boost or pull beyond that. It has a unified vapor chamber but rather small fans and overall cooling unit size. The Thinkpad P16 is a 115W A4500, with dynamic boost to 130W. It does have TWO separate vapor chambers one for CPU and another for GPU, so the best cooling and highest TDP. The stock paste is NOT Honeywell PTM79xx so the factory job can vary. The Precision 7770 was generally 120W from my experience, nothing I did could get it up to 130W where Aaron's machine would sit hence me thinking he had an early production machine that somehow accidentally has higher boost limits. This 7670 appears to be 90W with boost up to 102W. It is a lot better price / performance though. As a refurb with the 12950HX and 3080Ti and FHD for only $2100 before taxes its significantly cheaper than the 7770 config. I also found the 18W extra GPU TDP only brought roughly 10% more performance. Even more hilariously the fans on the 7670 do not have the stupid grinding/whoosh noise when cycling on/off.
-
I am pretty sure Dynamic Boost is working on the 7670 with 3080Ti. If I disable Nvidia PCF in the device manager, the sustained TDP is 90W, but if it is enabled then the sustained TDP is 102W. https://www.3dmark.com/compare/spy/33796418/spy/33791770#
-
I am unsure if the Lenovo A4500 module was some kind of MXM card, it does have an edge connector but obviously not the right form factor. The base TDP was 115W and dynamic boost was up to 130W.
-
Attached is the Thinkpad P16 A4500 vbios. p16a4500.zip
-
Yes. Dynamic Boost is available and enabled. I also tested it with Nvidia PCF disabled in the device manager, which then limited the GPU to 90W. So it does appear to get 12W through dynamic boost.
-
Indeed, glad to see this issue fixed even if there is no mention in the release notes. They technically also fixed the GT DC Loadline setting to be the proper 4.0 ohm setting, but that is more of an issue where load voltage for the iGPU is reported to be higher than the actual value ( ideally you want AC == DC ) I also tested Dynamic Boost 2.0 but did not see GPU TDP above 102W, or near 136W combined CPU+GPU under load. So that is still not working with current bios + ec + vbios + 517.66 dell driver.
-
Just a heads up, bios 1.8.0 fixes IA AC loadline values to the correct values. Tested this on a fresh machine running 1.6.3 from the factory.
-
I think the best way to be happy with the 3080ti option as a consumer purchase is to only consider it in a refurb model. If you are paying around 2000 then it's perfectly fine for it to perform in the lower bracket. Even with the lower performance it can run games well enough while being a business chassis.
-
Have you tried flashing Aaron's 3080Ti vbios to your machine? IIRC his was delivered pretty early on so it may have different base TDP or DB2.0 configuration settings. 130W pull should at least boost your score up a bit even if it cannot reach 150W.
-
Thank you for the information. It is interesting that the A4500, A5500 and 3080Ti are all around the same power limit and performance levels. Does afterburner allow manipulation of the voltage curve, overclocking, or power limit with any drivers? In that past I thought it was completed locked. If the setup works for you, using the Intel GPU to drive all displays definitely can reduce idle power draw. The Nvidia GPU will also be in a different higher power state depending on if you have many displays connected or if the displays are set to a higher refresh rate. Obviously be sure to have IA AC LoadLine set correctly if it is not otherwise CPU voltage will be significantly overcompensated.
-
By chance does anyone happen to know the sustained TDP for the 7670 A4500 GPU? From Notebookcheck's review I can see the 7670 A5500 is about 102W but wondering if the A4500 is the same.
-
They probably want to keep the performance jump just enough to entice upgrades and leave room on the table for future products to look better. If 5090 mobile architecture ends up not much more efficient they can just add more TDP to keep the performance jump. I think this is why the so heavily push 4k panels because with QHD / FHD the same GPU can remain relevant for much longer.
-
I do need to sell it. I can make you a good deal on the barebone if you are interested!
-
I have a 330W brick from an Alienware 51M R2. While it does recognize and power the Precision 7770, it did not allow draw over 240W. In fact there were still situations where combined full load use was causing battery drain. I would think that if the Dynamic Boost 2.0 limit is 150W then a synthetic load with CPU idle (<10W) and GPU 100% should be able to reach that value, but the highest I ever saw was 120W regardless of driver, bios and operating system combination.
-
Well HP definitely has their own custom GUI bios implementation for their business laptop line, similar to Dell and Lenovo. It is hard to say if it could use some optimization but they may not want to touch it so long as it is stable. I did not spend too much time in the bios on mine (maybe 2-3 minutes) and did not notice fan noise going above 2500rpm, other than when I performed the bios update where it always sets the fans to max speed. I have wondered if rendering the bios GUI to a 4k display makes it consume more CPU power. My Zbook Fury G9 had a WUXGA panel but my wife's Zbook X2 G4 tablet has a 4k panel and is noticeably louder and more laggy in the bios GUI.
-
So far as I know the Prema bios still isn't available. But you can modify hidden UEFI setup variables to get the same effect.
-
The system will consume more power in the bios because no drivers are loaded for the CPU and GPU to enable c-states which allow the chips to drop into lower power consumption modes based on demand. Also, the 12900HK has less cores, cache and no PCH so it could draw as much as 20W less in this scenario compared to 12900HX + WM690. The stock paste job can also have a chance to be poorly applied at the factory which can result in higher fan noise as the highest core temperature is usually used as the determination for fan speed. Check the P-core temperature deltas under load with CBR23 Multi as usually a poor paste job could have as much of a 20C difference between the highest and lowest. It is especially unlucky if the favored cores have poor contact because that means it likely will thermal throttle in single core loads and cannot reach a full 5.0ghz boost, along with crappy fan response as load spikes can cause 50C jumps.
-
I wonder if running the system with the stock bios AC LLC values, and resulting load voltages of 1.5v+ at 3.2ghz will cause premature chip failure. Its a pretty dramatic difference compared to a properly tuned Alder Lake chip which can be around 1.0v at 4.2ghz. I think the XOC crowd generally says over 1.4v sustained is bad without very good cooling in place. I thought last year's primary Gen4 drive slot deactivation switch flakyness was bad but this could be just as bad in terms of RMA headache.
-
Were you actually able to get 125W out of the 3080Ti in the 7670? I have found that dynamic boost 2.0 in general has not worked at all with the Precision 7770. The 3080Ti that should be 150W only pulled 120W, and the A3000 that should be 130W only pulls 115W, regardless of bios settings, performance mode, driver used, Nvidia PCF sourced from Dell drivers, plenty of thermal headroom.
-
The same technique applies across all the generations, just the version of ME Tools has to match your ME Firmware version: https://brendangreenley.com/undervolting-2020-dell-laptops-like-the-vostro-7500-and-more-tips-to-improve-thermals-battery-life-and-speed/#cpu-undervolt With the Thinkpad P16 I ran into an issue where the bios setup variable region was locked even in the grub pre-boot environment. Probably some security related setting that is not modifiable. Although it did not really need undervolting anyways as the cooling system (CPU side vapor chamber + PTM7950 + good AC/DC LLC config + 157W PL1/PL2) did well as is. You could probably expect 22000+ in CBR23 from the 12900HX as its the most setup to handle it as a turnkey solution.
-
I think Dell could have had an easier time marketing CAMM if they had made a unique module for the Precision, say a 32GB DDR5-5600 CL36 XMP which could work as it was commonly paired with a 12900k in a desktop. Although maybe the Alienware group did not want this given that the Precision already had HX CPU's, 3080Ti and a 120hz display option.
