Mobile Computing Forum - Other UMPC Hardware - Benchmark results & Battery life tests on eo.

UMPCPortal.com Forum Index

Other UMPC Hardware

Benchmark results & Battery life tests on eo.

Browsing this Thread: 1 Anonymous Users

Bottom

Previous Topic

Next Topic

Topic options

Search Forum

Search Forum

Advanced Search

View mode

View mode
Threaded
Newest First

Poster	Thread

ctitanic

Re: Benchmark results

#71

Joined: 2006/3/8

From

Posts: 680

Quote:

jtokash wrote:
Quote:

ctitanic wrote:
Quote:

jtokash wrote:
Also, using the util chippy suggested (cpuspeed), I'm seeing that my cpu idles at 400Mhz when connected to AC and 1ghz when not connected to AC. Wierd.

Well, that's weird. My CPU goes to 400 when it's idles no matter if I'm on battery or connected to power.

Ctitanic - hmmm... I double checked after a reboot and it's still happening for me.

Check your power squema select Laptop or Max Battery.

_________________
MY UMPC BLOG

Posted on: 2006/5/10 4:38

ctitanic

Re: Benchmark results

#72

Joined: 2006/3/8

From

Posts: 680

Quote:

Mr_g@dgeTT wrote the following post at 05-09-2006 10:40
PM:

For some reason I am having trouble posting over at Origamiportal all of the sudden.

I played solitaire for 20 minutes with the eo going in and out of C3 power state. All USB was disabled. Screen at 20-25% bright.

It started out at about 96% and 2:22 hours reported.

After 20 minutes of play, it reports 78% 1:55 hours.

kinda promising.....

thoughts?

This is ready good news!

_________________
MY UMPC BLOG

Posted on: 2006/5/10 4:45

jtokash

Re: Benchmark results

#73

UMPC Curious

Joined: 2006/4/23

From

Posts: 19

Quote:

ctitanic wrote:
Quote:

jtokash wrote:
Quote:

ctitanic wrote:
Quote:

jtokash wrote:
Also, using the util chippy suggested (cpuspeed), I'm seeing that my cpu idles at 400Mhz when connected to AC and 1ghz when not connected to AC. Wierd.

Well, that's weird. My CPU goes to 400 when it's idles no matter if I'm on battery or connected to power.

Ctitanic - hmmm... I double checked after a reboot and it's still happening for me.
Check your power squema select Laptop or Max Battery.

Max Battery works - gives me 400mhz at idle with AC or Battery. I guess the Portable/Laptop setting for me is broken.

Edit: THANKS Ctitanic!

_________________
--
http://blog.tokash.org/

Posted on: 2006/5/10 4:46

chippy

Re: Benchmark results

#74

Site Administrator

Joined: 2006/3/2

From Bonn, Germany

Posts: 3727

I missed all the action!

Have just read through all the posts and it looks like we're getting some answers on the C3 issue. It is the USB bus causing the problems
Ctitanic reports that someone is seeing 2:25 at low-use now. Its a lot better than 1:30. In fact, a little more than expected. Power-saving will only get you about 4W power reduction. Thats 13W total drain which equates to exactly two hours. We'll have to wait for more tests to come in to double check the 2:25 figure.

13W is still 30% too high though and indicates that either the battery is not 26whr or theres a high-drain component in there. Possibly the screen.

If someone can run idle tests with screen full, screen min and screen off (is that possible?) then we can work out the screen drain.

Also, someone needs to double-check the current draw from the battery at idle and at load.with a mumtimeter between the battery + or - and the battery connector if you can work out which contact it's on and wire it up! Be careful with that test though.

With those two tests we'll be able to works out if the battery is 26W/hr as stated or whether the screen is a high-power drain item.

Well done to all who tested this out. I hope Tablet Kiosk, Paceblade and Amtek have been watching and are now kicking the designer as we speak becuase not being able to use low-power CPU states on a VIA chipset is a bad design and a serious waste. FIngers crossed that this can be fixed in software.

Regards
Steve.

_________________
Dont forget to check out the LIVE page for live chat.

Posted on: 2006/5/10 8:42

ctitanic

Re: Benchmark results

#75

Joined: 2006/3/8

From

Posts: 680

I'm back at work so I can't run tests until I get home in about 8 hours but I want to add that would me good if mgagett can run a few tests but instead of disabling all the USB listed, just disabling one by one and checking the C3 stage. Probably is just one of them the one causing the problem? Don't you think?

_________________
MY UMPC BLOG

Posted on: 2006/5/10 13:27

chippy

Re: Benchmark results

#76

Site Administrator

Joined: 2006/3/2

From Bonn, Germany

Posts: 3727

Good idea Ctitanic.

We'll have to wait until the rest of you US guys wake up I think.
Roy from Oprigamiportal only went to bed a few hours ago!!

Would you have a chance to do the screen-off idle test?
I don't know how you would completely turn the screen off though!

You said that PowerDVD didnt work for you. When John Tokash ran the test he didnt mention any jerkyness.

S.

_________________
Dont forget to check out the LIVE page for live chat.

Posted on: 2006/5/10 13:51

mgadgett

Re: Benchmark results

#77

UMPC Curious

Joined: 2006/5/10

From

Posts: 11

Quote:

ctitanic wrote:
I'm back at work so I can't run tests until I get home in about 8 hours but I want to add that would me good if mgagett can run a few tests but instead of disabling all the USB listed, just disabling one by one and checking the C3 stage. Probably is just one of them the one causing the problem? Don't you think?

Unfortunately, I tried the disabling USB one at a time last night, in various combination, and the PerfMon C3 chart didn't start spiking until the very last USB was disabled (then if I re-enabled any others the C3 stayed at 0%).

So... you have to disable all.

Posted on: 2006/5/10 13:54

chippy

Re: Benchmark results

#78

Site Administrator

Joined: 2006/3/2

From Bonn, Germany

Posts: 3727

Hi Mgadget.

The device you disabled last - did you try disabling it on its own too?

Has anyone got any decisive figures on idle battery life now withouth USB? I've seen guestimates between 2 and 2.25 hours.

Steve

_________________
Dont forget to check out the LIVE page for live chat.

Posted on: 2006/5/10 14:14

mgadgett

Re: Benchmark results

#79

UMPC Curious

Joined: 2006/5/10

From

Posts: 11

Quote:

chippy wrote:
Hi Mgadget.

The device you disabled last - did you try disabling it on its own too?

Has anyone got any decisive figures on idle battery life now withouth USB? I've seen guestimates between 2 and 2.25 hours.

Steve

Yeah... tried that as well.

I sat there for 30 minutes trying loads of USB disable/enable combination. I only had luck when all of the USB were disabled.

I had to stop the battery life test while playing Solitaire last night to get some sleep. Check my post over at Origamiproject. I placed the device in Standby and will resume late this evening. In the meantime, anyone else feel free to run some various battery tests once you get C3 states.

I'll see if I can dig up anything on 7" color LCDs. I am still hopefully for around 2 hours with Solitaire.

BTW... my unit has the 5400RPM 40GB hard drive and teh 512MB RAM. I will post datasheet links for both those devices later.

-mgadgett/Mr_g@dgeTT

Posted on: 2006/5/10 14:41

chippy

Re: Benchmark results

#80

Site Administrator

Joined: 2006/3/2

From Bonn, Germany

Posts: 3727

Mgadget.

Sorry. I had to ask the question!!!

Thanks for looking for the other info.
2hrs with Solitare is still a bad figure though.
I'm really hoping someone can test the battery current.
Marix said he'd try to get a multimeter today so i've got my fingers crossed.

I wish I could help on the practical side but there's no MS UMPC's in Germany yet. Could be even longer if we uncover any more problems like this!!

Steve.

_________________
Dont forget to check out the LIVE page for live chat.

Posted on: 2006/5/10 14:49

mgadgett

Re: Benchmark results

#81

UMPC Curious

Joined: 2006/5/10

From

Posts: 11

If someone does monitor the current from the Lithium-Ion battery, PLEASE, PLEASE, PLEASE... be very careful!

Lithium batteries SHOULD have built in current limiting circuitry to prevent the battery from overheating and exploding if you accidentally short the positive and negative, however, some manufacturers cheap out on putting in good protect to lowball others on the bid. (hence the many recalls on exploding cellphone batteries. We have even done some destructive testing on lithiums in our lab at work).

Not to scare anyone... as this test can be done safely. Just wear some safety glasses and use common sense.

Posted on: 2006/5/10 14:58

ctitanic

Re: Benchmark results

#82

Joined: 2006/3/8

From

Posts: 680

I have not had a chance to play around to see if it's possible to have the eo running (lets say playing music) with the screen off.

mgadgett tests about battery life are good but I still think that we need more tests there. When I get home today I'll disable all the USB devices and play a movie in that way to see how much battery I get in this mode.

Any way guys, we have done a very good job here and I really believe that we have found what was the problem. Here is what I just posted at my blog: Quote:

Geek.com published some time ago a review of a laptop using the same VIA processor that we have in the eo V7110. Here is what the author wrote about what is going on with battery life on VIA processors:

I should note here that the TongFang (this is the Laptop that they were reviewing) does not enter C3/C4 power states; that fact was revealed to me by a BIOS author at VIA Technologies. I'm told this is actually a common problem on computers using USB 2.0 devices because constant polling actually causes the processor to frequently enter a C0 state (the least amount of power savings) while "talking" with certain South Bridges. This relates back to the stories I've heard lately about Intel's new Core Duo processors having an unusually high amount of power drain due to a bad driver, as was originally reported by Tom's Hardware. Apparently this is something the industry has known for quite some time but has been quiet about. Since I wasn't using any USB devices on the TongFang during the power tests, I suspect the SD card reader was the culprit here, keeping the notebook out of C3/C4 levels of power savings, which are the greatest power saving modes.

Without C3/C4 power states available to it (due to the known USB-induced "bug"), the TongFang is not able to come close to revealing what the C7-M is capable of delivering in terms of battery life longevity. VIA claims to have seen battery life up to 8 hours with notebooks that fully exploit the C7-M's power-saving technology. VIA has told me that it's working on a BIOS fix, but at the time of this writing I had not yet received said fix.

This is exactly what we have found in our tests described at OrigamiPortal Forum.

TabletKiosk/AMTek will have to work hard on this issue and demand a fix from VIA as soon as possible. I love the eo V7110 but really its battery life is just a joke that will damage significantly the amount of sales of these units here in USA where the Q1 as began with some very good reports of more than 2 hours of battery life while browsing the web and doing normal activities. 1 hour and 25 minutes is one of the worse battery life that I have seen in a long time, I can't consider it even close to a poor battery life (2 hours in any laptop) and it's far away from my concept of "mobility". I consider the external design of the eo a lot more practical than what I see in the Q1, but a better battery life will be something that many customers will appreciate a lot more than external design.

The good news is that all this seems to me that can be fixed with a software patch via a BIOS update or just a driver update.

Another good news is that yesterday I sent TabletKiosk all this information pointing them to the thread at OrigamiPortal forum where we have posted all the information that we have been able to find and test. I hope that that will help them to find a fast solution for this issue.

--
Posted by CTitanic to ULTRA MOBILE PCs TIPS at 5/10/2006 09:00:41 AM

_________________
MY UMPC BLOG

Posted on: 2006/5/10 15:07

ctitanic

Re: Benchmark results

#83

Joined: 2006/3/8

From

Posts: 680

TabletKiosk knows about what is going on. I opened a Ticket with them and forwarded them the whole information with a link to this thread. I received two confirmations from them, one to tell me that all the info has been forwarded to their engineers and the other to tell me that the info was forwarded to Martin Skemal (CEO of the Company).

_________________
MY UMPC BLOG

Posted on: 2006/5/10 16:02