Temash Tubo Dock demo at MWC

[Maragark]

I happened to notice a engadget article when looking for Temash benchmarks.

The Temash based tablet could show 80 fish at 60fps, the Clovertrail based tablet could show 1 fish at 30fps.

Does anyone have a Clovertrail based tablet? If so, could you run the HTML5 Fishbowl in IE and check the performance?

 

[TuxDave]

My Core i5 laptop in Firefox doesn't even do that well on that Fish benchmark. What kind of benchmark is this???

http://ie.microsoft.com/testdrive/performance/fishbowl/

 

[ShintaiDK]

HTML5 performance depends on browser as well.

Examples:

Never trust PR stunts.

 

[sm625]

I would be surprised if it could hit 30fps with 1 fish. Though there isnt much difference between 1 fish and 100 fish. Just turning off the water and shadows and that other stuff makes a bigger difference.

 

[Olikan]

M$ benchmark, IE10 runs extremily well

seems legit and fair :hmm:

 

[ShintaiDK]

M$ benchmark, IE10 runs extremily well

seems legit and fair :hmm:

To add to the stupidity. I just upgarded from IE9 to IE10 on my Win7. While I use vsync, so bare with me. But 2000 fish went from 30FPS to 60FPS.

 

[Gikaseixas]

Benchmarking apart, this seems to be an interesting option to what's available right now. If they can make it fast enough for some light gaming I'm in.

 

[Abwx]

HTML5 performance depends on browser as well.

Examples:


Never trust PR stunts.

What this has to do with the claimed numbers ?...

Does this chart makes them irrelevant or wrong?...

That s just trolling all the way.

About the said bench :

Microsoft’s Fish Bowl HTML5 test for Internet Explorer managed 60 fps with the maximum number of fish. When docked, the fishbowl score rose 40-50%.

http://news.softpedia.com/news/MWC-2013-AMD-Turbo-Dock-Demoed-on-Temash-Tablet-332291.shtml

 

[Gideon]

http://news.softpedia.com/news/MWC-2013-AMD-Turbo-Dock-Demoed-on-Temash-Tablet-332291.shtml

The softpedia article only references another one, which is much better:

http://www.slashgear.com/amd-turbo-dock-temash-tablet-reference-design-hands-on-24271087/

I also already linked this a while ago here:

http://forums.anandtech.com/showpost.php?p=34672974&postcount=1523

 

[Maragark]

I would be surprised if it could hit 30fps with 1 fish. Though there isnt much difference between 1 fish and 100 fish. Just turning off the water and shadows and that other stuff makes a bigger difference.

I just rechecked the video and all the bells and whistles were enabled. I just checked the specs for the Atom Z2760 and its GPU only supports DX9.3. Given that the fishbowl test only works in IE, it's likely using Direct2D which requires DX10.

 

[ShintaiDK]

I just rechecked the video and all the bells and whistles were enabled. I just checked the specs for the Atom Z2760 and its GPU only supports DX9.3. Given that the fishbowl test only works in IE, it's likely using Direct2D which requires DX10.

And any Direct2D and DirectWrite would be emulated via the CPU for DX10 if the GPU is DX9. (With platform update.)

 

[TuxDave]

And any Direct2D and DirectWrite would be emulated via the CPU for DX10 if the GPU is DX9. (With platform update.)

Oh marketing... show them a way to skew a benchmark or give them some fancy sounding technical term, they'll ride it all the way to the bank.

 

[AtenRa]

The Intel Z2760(clovertrail) has the PowerVR SGX 545 which according to Imagination is DX-10.1 capable.

 

[Exophase]

I said it in the other thread but it really bears repeating..

You can't do a benchmark comparison between two devices by changing both the load AND the allowed performance. You can't solve for two unknowns in a system of one equation.

Either you fix the load and measure the performance (easy) or you fix the performance and vary the load (hard). Since performance measurements tends to be linearly proportional with capability that makes more sense anyway, there's not necessarily going to be an easy mapping between load and capability.

In other words, keep the damn fish the same.

(this is also directly related to my constant complaint that comparing peak perf vs peak power consumption between two platforms is totally unfair if you don't keep one of the two things fixed, but people keep doing that too)

 

[Vesku]

Hmm, I ran the fishbowl in Chrome and was at 10000+ fish 60fps with Radeon 7950 kept climbing slowly but got tired of watching it.

 

[ShintaiDK]

The Intel Z2760(clovertrail) has the PowerVR SGX 545 which according to Imagination is DX-10.1 capable.

Besides this:

The Intel Graphics Media Accelerator 3600 or GMA 3600 is an integrated processor graphics card (e.g. in the Atom N2600) without dedicated memory. It is based on a PowerVR SGX 545 design similar to the GMA 500 (SGX 535). In theory the GPU should support DirectX 10.1 but at launch only DirectX 9 support was available.

I dont think thats changed since the same applies for the ATIV Smart PC XE500T1C-A02DE thats Clover Trail based.

 

[pm]

My Core i5 laptop in Firefox doesn't even do that well on that Fish benchmark. What kind of benchmark is this???

http://ie.microsoft.com/testdrive/performance/fishbowl/

For what it's worth, I'm using Chrome 25.0.1364.97m on a Core i5-based Lenovo 520. On the page you linked, I get 60fps with all layers checked until 50 fish, then drop into the high 50's for 50 fish, high 40's for 100 fish, mid-30's for 250 fish, 20's for 500 fish and lastly 12fps for 1000 fish.

 

[Maragark]

The Intel Z2760(clovertrail) has the PowerVR SGX 545 which according to Imagination is DX-10.1 capable.

I took the info from Anand's review:

"The GPU is a PowerVR SGX 545 from Imagination Technologies, clocked at a very high 533MHz. Direct3D feature level 9_3 is officially supported."

It just an assumption, but it seems the most likely explanation so far.

Edit: It's no longer an assumption, Intel states quite clearly in the datasheet for the Atom Z2760 that, "Support DirectX* 9.3 compliant Pixel Shader* v2.0 and OpenGL* 2.1"

 

[Ventanni]

not an amd chestbeater.

it's a silly benchmark, but there should be no question that the gpu in temash is vastly superior to the gpu in a comparable atom platform. if anything.it shows better driver support on amds part. silvermont should be a better comparison though

 

[ShintaiDK]

It might not be driver related, but hardware instead. I think there is a reason why Intel is going to use their "HD" GPUs in the next Atoms.

 

[Exophase]

It might not be driver related, but hardware instead. I think there is a reason why Intel is going to use their "HD" GPUs in the next Atoms.

You think IMG not only lies about the hardware capabilities of their IP but was able to fool a licensee as experienced as Intel for so long that they brought it to market?

Intel themselves advertized their SGX545 parts as DX10. I think they'd know if it was possible or not. They only changed it later, which sounds much more likely to be driver related.

Intel has lots of incentive to use their own in-house GPUs, although word on the street is that PowerVR cores aren't going anywhere for the lowest power phone SoCs.

 

[Abwx]

Anyway , a quadcore powered tablet with DX11 capability
is appealing for tablet lovers but the most interesting is its
docking possibility that will attract notebooks users and
shift even more a part of this market in favour of such
convertibles devices.

 

[Blandge]

Anyway , a quadcore powered tablet with DX11 capability
is appealing for tablet lovers but the most interesting is its
docking possibility that will attract notebooks users and
shift even more a part of this market in favour of such
convertibles devices.

I don't see what's so revolutionary about lowering max cpu frequency when not attached to docking station. I guess it's just good marketting from AMD for once.

 

[TuxDave]

I don't see what's so revolutionary about lowering max cpu frequency when not attached to docking station. I guess it's just good marketting from AMD for once.

It's about raising TDP when you're docked. That's the best part of the whole thing and what I'm waiting for.

 

[LogOver]

i5-2410M (HD3000), Chrome 25.0.1364.97
500 fish - 60 fps
1000 fish - 32 fps
2000 fish - 14 fps
Similar result in FF19.