Welcome to our peer-to-peer forums, where owners help owners. Need help now? Visit eSupport here.

English Community

Linux Operating SystemsOther Linux Discussions
All Forum Topics
Options

389 Posts

03-06-2021

Germany

219 Signins

2345 Page Views

  • Posts: 389
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2345
  • Message 381 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-06-20, 16:25 PM

@ le-dvd wrote:

It' been months since we ask what we can do to help.

 

I understand that at one point people begin losing their patience. 

These machines cost a lot, and as a customer you don' really care who or how an issue has to be fixed.

 

I' sure Mark does his best, and there' a lot of people in this thread willing to help (we can share data, we can test firmwares, even sending our laptop if needed), but it doesn' seem to make anything better or faster.So at this point, what more can we do ? Wait a few more months ?

 

Unfortunately, we can do nothing.

 

I offered myself as a beta tester or whatever is needed in various threads for various issues,

including this one. I would even sign NDA's :).

 

I looked myself at ACPI dumps for Linux/Windows mode but unfortunately, Lenovo has some unusual

interfaces, and way too many 'magic numbers'. I figured some registers but without docs, there is no way of really knowing.

Also, there is more to look at, like AGESA code/implementation and much more we don't have access to.

 

 

 

 

Reply
Options

10 Posts

05-18-2021

United States of America

26 Signins

105 Page Views

  • Posts: 10
  • Registered: ‎05-18-2021
  • Location: United States of America
  • Views: 105
  • Message 382 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-06-20, 16:43 PM

There isn't a standard for 'power drain on mode',

 

@osnix , thanks for the feedback. You may be right overall (will explain below), but you are wrong regarding the standards for energy consumption

 

Both X13 AMD and T14s Intel are Energy Star certified

https://www.energystar.gov/productfinder/product/certified-computers/details/2359937 (X13 AMD)

https://www.energystar.gov/productfinder/product/certified-computers/details/2355137 (T14s Intel)

 

According to the tables, X13 AMD consumes 25-30% more in power-off and S3 states than the T14s Intel.

 

Whether X13 AMD in absolute numbers still fits the specs from the table I don't know, but I'm 100% confident, that relatively to the T14s Intel my X13 AMD is one order of magnitude worse (>1000%, not nearly 25%) than my T14s Intel.

 

So 2 options here:

1. either the numbers in the table allow for a generous headroom, and both Intel and AMD match the specs (but Intel has plenty of headroom); or

2. AMD, being much worse than Intel, doesn't indeed fit those numbers

 

I believe it's (1) however, that's why you may have a point overall about the absence of the case here.

 

Reply
Options

389 Posts

03-06-2021

Germany

219 Signins

2345 Page Views

  • Posts: 389
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2345
  • Message 383 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-06-20, 16:57 PM

@ koiuo wrote:

There isn' a standard for ' drain on <some supend> mode',

 

@osnix, thanks for the feedback. You may be right overall (will explain below), but you are wrong regarding the standards for energy consumption

 

Both X13 AMD and T14s Intel are Energy Star certified

https://www.energystar.gov/productfinder/product/certified-computers/details/2359937 (X13 AMD)

https://www.energystar.gov/productfinder/product/certified-computers/details/2355137 (T14s Intel)

 

According to the table, X13 AMD consumes 25-30% more in power-off and S3 states than the T14s Intel.

 

Whether X13 AMD in absolute numbers still fits the specs from the table I don' know, but I' 100% confident, that relatively to the T14s Intel my X13 AMD is one order of magnitude worse (>1000%, not nearly 25%) than my T14s Intel.

 

So 2 options here:

1. either the numbers in the table allow for a generous headroom, and both Intel and AMD match the specs (but Intel has plenty of headroom); or

2. AMD, being much worse than Intel, doesn' indeed fit those numbers

 

I believe it' (1) however, that' why you may have a point overall about the absence of the case here.

 

 

Yes, but it is not an Industrie standard at all :).

 

Nothing says: Platform X, with Processor Z, gets only certified if power drain is less that..., in modeZ

 

'Energy efficient' can mean 'whatever' in this case.

 

Also, you forgot:

 

3) BIOS & platform integrators didn't yet optimize for AMD platforms.

 

Which, when you follow, kernel changes regarding that, and therefore, BIOS changes for other

Vendors are more likely the issues here.

 

 

Reply
Options

684 Posts

07-22-2019

Canada

475 Signins

5426 Page Views

  • Posts: 684
  • Registered: ‎07-22-2019
  • Location: Canada
  • Views: 5426
  • Message 384 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-06-21, 21:32 PM

Sorry for the slow reply - I know I'm overdue for an update. Been holding off a little bit as I don't really have anything news worthy and I'm waiting on some results.

 

There was a goof-up communications-wise between myself and the team in China, partly my fault, that delayed things a bit (there was a question that I thought was a statement and didn't realise they were waiting for a reply). They are re-doing the energy cert with Linux so that we can compare official measured numbers with Windows. It's best to get the official measurements to work from as it's what we ultimately have to target.

 

One thing I wasn't 100% sure above from the thread above is if the sleep battery issue is being seen in Windows as well or if it's just Linux. If we're the same as Windows  I'll still keep digging to improve it with the AMD team but it will make it a harder exercise.

 

For e-cert, the test is the same regardless of CPU so whether it's AMD or Intel it still has to meet the same specs. From my experience it's non-trivial passing the tests - we've had more than our fair share of failures on Linux and the sources of failure can be multiple and complicated (1st place goes to the AMD platforms this year for sure ...).

If you're interested the spec is online: https://www.energystar.gov/sites/default/files/ENERGY%20STAR%20Computers%20Final%20Version%208.0%20Specification%20-%20Rev.%20April%202020.pdf

The e-cert is based on four different measurements leading to an eTEC value that determines pass or fail - page 12/13

 

As an aside - there are another 6 energy certs we do for different regions, but e-star is the main one and usually the hardest to pass.

 

Mark

 

 

Reply
Options

83 Posts

04-17-2021

Greece

59 Signins

380 Page Views

  • Posts: 83
  • Registered: ‎04-17-2021
  • Location: Greece
  • Views: 380
  • Message 385 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-06-21, 21:52 PM

They are doing energy cert things fo a year now...

 

It is like Valve and Half Life...

 

It seems to me that they are clueless of the issue. Should we start a fundraiser and send them two T14 laptops? One with windows and one with GNU/Linux with instructions how to put both in sleep?

 

They will keep the energy cert thing until the heat death of the cosmos.

Reply
Options

389 Posts

03-06-2021

Germany

219 Signins

2345 Page Views

  • Posts: 389
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2345
  • Message 386 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-06-21, 23:13 PM

@ MarkRHPearson wrote:

.....

For e-cert, the test is the same regardless of CPU so whether it' AMD or Intel it still has to meet the same specs.

 

This is the first mistake they are doing, at least in my opinion ( and after reading some of their formulas )...

And then, they are calculating idle modes, power off, ram, etc following  'IEC 62623, Ed.1.0, 2012-10'... Uh Oh...

 

Anyway, regarding idle modes, power off, etc e-cert is useless.

As long the overall results following the formula(s)  are within their percentage everything is fine for them.

 

Reply
Options

49 Posts

09-04-2010

netherland

203 Signins

1275 Page Views

  • Posts: 49
  • Registered: ‎09-04-2010
  • Location: netherland
  • Views: 1275
  • Message 387 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-06-22, 7:11 AM

Maybe it is possible to bring the whole thing under attention of some Linux magazines. (e.g. Linux Format) and so on.

Create a bit of awareness of all the current problems and that their 'certification' not is what customers expected to be.

YouTube is also a good place to make people aware of issues with Linux, and advise them to go for an older model.

 

 

 

Reply
Options

8 Posts

06-22-2021

Germany

10 Signins

50 Page Views

  • Posts: 8
  • Registered: ‎06-22-2021
  • Location: Germany
  • Views: 50
  • Message 388 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-06-22, 17:00 PM

@MarkRHPearson 

Regarding your question, whether the sleep-drain occurs in Windows and Linux:

 

In my experience (i have dual-boot manjaro and Win10) the massive sleep-drain only occurs in Linux. However i dont have any exact numbers.

Reply
Options

45 Posts

10-11-2020

United States of America

42 Signins

410 Page Views

  • Posts: 45
  • Registered: ‎10-11-2020
  • Location: United States of America
  • Views: 410
  • Message 389 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-06-22, 17:59 PM

@ MarkRHPearson wrote:

 

One thing I wasn' 100% sure above from the thread above is if the sleep battery issue is being seen in Windows as well or if it' just Linux. If we' the same as Windows  I' still keep digging to improve it with the AMD team but it will make it a harder exercise.

 

 

Here's my thread about s3 from win10. Why does @someotherguy say it's unsupported? The hardware supports s3, win10 supports s3.

Reply
Options

389 Posts

03-06-2021

Germany

219 Signins

2345 Page Views

  • Posts: 389
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2345
  • Message 390 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-06-22, 18:10 PM

@ t14g1amd wrote:

@ MarkRHPearson wrote:

 

One thing I wasn' 100% sure above from the thread above is if the sleep battery issue is being seen in Windows as well or if it' just Linux. If we' the same as Windows  I' still keep digging to improve it with the AMD team but it will make it a harder exercise.

 

 

Here' my thread about s3 from win10. Why does @someotherguy say it' unsupported? The hardware supports s3, win10 supports s3.

 

That is a very good question.. 

 

I guess the same problem will pop up once Linux fully supports S0ix...

The thing is, S3 isn't going to disappear in Linux, nor will it disappear in W10,

even is off by default in W10, is still supported by Windows.

 

Always the same story with those companies ;)

Reply
Forum Home

Community Guidelines

Please review our Guidelines before posting.

Learn More

Check out current deals!

Go Shop
X

Save

X

Delete

X

No, I don’t want to share ideas Yes, I agree to these terms