• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

What is your HTML performance 2D

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Brunnis

Senior member
Nov 15, 2004
501
56
91
Could anyone with a HD3000 run this test? Would be very interested if they get the same poor result.
MacBook Air with Core i7-2677M (1.8GHz, HD3000): 6 seconds, using FF12

So, it appears the HD3000 is okay for this kind of work.

EDIT: Test was run in Windows 7 64-bit.
 

lopri

Elite Member
Jul 27, 2002
12,842
325
126
I just reduced the window dimensions and it seems to be running slightly faster (the average draw duration is now closer to 200 ms). It's still running at 4 FPS though, and bogging down the entire system.

EDIT:

2728 seconds (almost 46 minutes)

And I cheated by reducing the window dimensions to make it run faster, too.
Hope you didn't stare at it while it's running. :twisted::awe:
 

lopri

Elite Member
Jul 27, 2002
12,842
325
126
I think this is more of a browser test than a hardware test. (except maybe G965 above) There are other HTML5 demos that are more suited for CPU or GPU.
 

2is

Diamond Member
Apr 8, 2012
4,281
130
106
I think this is more of a browser test than a hardware test. (except maybe G965 above) There are other HTML5 demos that are more suited for CPU or GPU.
Well I've used IE9 on all my machines (except the Mac) and have had varying results.

3770K + GTX680: 6 seconds
3770K + HD4000: 6 seconds
i3 2120 + HD 2000: 10 seconds
C2D e7500 + G41: 13 seconds

The differences may not SEEM like much but if you break it down to percentages it's huge.
 

Jeff7

Lifer
Jan 4, 2001
41,599
17
81
It means you either have a ultra crappy GPU (probably not the case) or a browser that does not support gpu acceleration or has it disabled.
Ah, that's probably it then. I think I had hardware acceleration disabled.

At home, with it enabled, it completed in 7 seconds. GPU usage spiked at around 30%. It's pegged at 60fps, maybe I'm vsync limited. Disabling, and restarting Firefox.

Aaaand nope. Changing vsync settings or refresh rate didn't let it go past 60fps.
(i5 750 @ 2.66GHz, Win7, 4GB RAM, ATi HD 5850)


In the fishbowl, 2000 fish drops me down to 23fps, and puts my GPU usage at around 80%.
 
Last edited:

beginner99

Diamond Member
Jun 2, 2009
4,861
1,255
136
10 seconds on a stock i3 2120 using the IGP (HD 2000)
So something is very wrong with my notebook...strange because the result was consistent across browsers and Intel driver versions. Maybe something with power settings? also it seems it it would have such a huge impact...
 

beginner99

Diamond Member
Jun 2, 2009
4,861
1,255
136
Ok, it was Power settings:

Set everything to max performance and woops: 6 secs.
The effect is also extremely noticeable in the fishbowl test.

Never thought those settings have such a huge impact.
 

rh71

No Lifer
Aug 28, 2001
52,728
916
126
wow Opera 11.64 failed at that miserably. IE9 was quick like everyone else.
 

BrightCandle

Diamond Member
Mar 15, 2007
4,762
0
76
Ok, it was Power settings:

Set everything to max performance and woops: 6 secs.
The effect is also extremely noticeable in the fishbowl test.

Never thought those settings have such a huge impact.
They have a massive impact. You may notice that in max performance Windows is performing significantly better as well. Current power saving while nice and efficient has a large and meaningful impact on the actual performance you see when using the computer.
 

hdfxst

Senior member
May 13, 2009
851
3
81
chrome-6 seconds,IE10-7 seconds and firefox-14 seconds and the fish bowl was chrome-28,IE10-42 and firefox-12.This was on windows 8
 

2is

Diamond Member
Apr 8, 2012
4,281
130
106
This test makes me wish I still had a Socket A system laying around so I can run it with my Radeon 9800 Pro that I still have around in a box somewhere.
 

ASK THE COMMUNITY