I’ve been around testing various forum softwares during the years, but since I’m not actively hosting any well visited forum I’m mostly interested in simplicity should the need arise. Amongst the most simplest solutions I’ve found is the Talki-forum plugin for WordPress: http://tal.ki/
So, I added exactly this forum to my site, it can be found under the “pages” button above in the navigation bar. If you should happen to have anything you would like to discuss, why not start a thread there.
First the run from user PytonOrm at sweclockers.com:
And then the run of exactly the same thing on my own rig:
The Ivy Bridge is slightly faster, but not much, but it is WAY hotter!
I don’t think I ever have to contemplate Ivy Bridge as an upgrade, rather a sidegrade or even thermal downgrade. Haswell will be the next interesting cpu, in 2013, until then I will be perfectly happy with using yesterdays news 
(IB-E could be interesting if it arrives at all)
AS my own sample of the Core i7 2600K is able to run at 5.0GHz+ speeds with a little vcore-bump, I decided to see if exactly that could outperform the brand new Core i7 3770K @ it’s approximate maximum possible OC of ca. 4.7GHz on air or entry level cloosed loop water cooling. It could. My last generation cpu can actually run faster than what is possible with the spanking new 3770K! I score 9.66p CPU in Cinebench R11.5 while the 3770K only reaches 9.52p CPU. Not a very big difference, but the new cpu is actually slower if one allows both of them to stretch it’s muscles. First the 3770K:
And then my own personal run:
Of notice is that the GPU was 99% loaded during the benchmark and yet the fan on the card never exceeded 57% fan speed, meaning it was never louder than my case fans. Conclusion is that an third party cooler is totally unnecessary for the GTX680, even with gpu speeds of 1200MHz+. And, I didn’t have to up the core voltage a single bit and there was no artefacts whatsoever. It seems I got me an overclocking beast
Find it here:
https://developer.spotify.com/technologies/spotify-play-button/
I decided to test it first with an awesome modern progressive rock band, Magic Pie:
Of course it only works for those who have Spotify installed, if on a handheld or a PC doesn’t matter, very cool!
A significant raise in 3D-points using 3DMark Vantage as well, up from ca. 22k to 34560P! It seems the card works very well right upon first run. Did I mention it is a lot more silent in operation than the GTX470. Full results below:
NVIDIA GeForce GTX 680 video card 3DMark Vantage benchmark test score
Processor
Processor
Processor clock
4789 MHz
Physical / logical processors
1 / 8
# of cores
4
Graphics Card
Graphics Card
Vendor
NVidia Corporation
# of cards
1
SLI / CrossFire
Off
Memory
2048 MB
Core clock
705 MHz
Memory clock
3004 MHz
Driver name
Driver version
8.17.13.110
Driver status
FM Approved
General
Operating system
64-bit Windows 7 (6.1.7601)
Motherboard
MSI Z68A-GD55 (G3) (MS-7681)
Memory
16384 MB
Module 1
4096 MB Kingston 9 @ 667 MHz
Module 2
4096 MB Kingston 9 @ 667 MHz
Module 3
4096 MB Kingston 9 @ 667 MHz
Module 4
4096 MB Kingston 9 @ 667 MHz
Hard drive model
120GB Corsair Force 3 SSD
Detailed scores
3DMark Score
34560 3DMarks
Graphics Score
35430
CPU Score
32189
Jane Nash
108.27 FPS
New Calico
99.14 FPS
AI Test
4488 operations/s
Physics Test
41 operations/s
Settings
Default settings used
Yes
Preset
Performance
Width
1280
Height
1024
MSAA
1
Texture Filtering
Optimal
Anisotropy
1
Texture Quality
Performance
Shader Quality
Performance
Shadow Shader Quality
Performance
Shadow Resolution Quality
Performance
Post Processing Scale
1:2
PPU Disabled
true
PPU Used
false
Disable Bloom
false
Disable Streaks
false
Disable Anamorphic Flare
false
Disable Lens Flare
false
Disable Lenticular Halo
false
Disable Motion Blur
false
Disable DOF
false
Disable Fog
false
Disable Color Noise
false
Flush On Low FPS
false
Fixed Framerate
false
GPU Count
1
Just changed card in my rig from a two year old Gigabyte GTX470, practically enough benchmarked in the previous post, to a MSI GTX680. The complete 3DMark11 score below:
NVIDIA GeForce GTX 680 video card 3DMark 11 benchmark test score
Processor
Processor
Processor clock
1596 MHz
Physical / logical processors
1 / 8
# of cores
4
Graphics Card
Graphics Card
Vendor
NVidia Corporation
# of cards
1
SLI / CrossFire
Off
Memory
2048 MB
Core clock
705 MHz
Memory clock
3004 MHz
Driver name
Driver version
8.17.13.110
Driver status
FM Approved
General
Operating system
64-bit Windows 7 (6.1.7601)
Motherboard
MSI Z68A-GD55 (G3) (MS-7681)
Memory
16384 MB
Module 1
4096 MB Kingston 9 @ 667 MHz
Module 2
4096 MB Kingston 9 @ 667 MHz
Module 3
4096 MB Kingston 9 @ 667 MHz
Module 4
4096 MB Kingston 9 @ 667 MHz
Hard drive model
120GB Corsair Force 3 SSD
Detailed scores
3DMark Score
P9709
Graphics Score
9847
Physics Score
10815
Combined Score
7720
GraphicsTest1
45.8 FPS
GraphicsTest2
45.98 FPS
GraphicsTest3
60.95 FPS
GraphicsTest4
29.91 FPS
PhysicsTest
34.33 FPS
CombinedTest
35.91 FPS
Settings
Default settings used
Yes
Edition
Basic
Dxgi Adapter Luid
0
Screen Width
1280
Screen Height
720
Msaa Sample Count
1
Texture Filtering Mode
Trilinear
Max Af Anisotropy
1
Tessellation Detail
5
Max Tessellation Factor
10
ShadowMapSize
5
Shadow Cascade Count
4
Surface Shadow Sample Count
16
Volumetric Illumination Quality
5
Ambient Occlusion Quality
5
Depth Of Field Quality
5
Enable Window Mode
Off
Enable Vertical Sync
Off
Enable Triple Buffering
Off
Enable Wireframe
Off
Color Saturation
100%
System Info Guid
D42FDD6E-A380-4BD8-9DD5-424566C937E9
So AMD and Nvidia has released new high-end videocards, HD7970 and GTX680, I’m sure you all know about them. They may be fine and awesome and all for multiscreen gaming in extreme resolutions like 5760x, but at the end of the day if one isn’t going to game at any higher setting than 1080P(1920×1080) most games won’t see any other benefit with these new cards than lower power draw and possibly lower noise. It’s doubtful if that is worth 500€ if one already posseses a highend end card from the last two years. If not, then it may be, even if expensive.
Take myself as an example, owning a 2 year old GTX470 and running my current favorite game Skyrim at it’s aboslute highest settings in 1080P and with the official high-resolution texture pack installed and running. Granted, I have to overclock the card a bit, but that ain’t to hard usings MSI:s Afterburner program. What I end up with is a constant steady 60fps during various gameplay, indoors and outdoors, and with multiple enemies. No dips in framerate at all, none, zero, nill! The only thing I would benefit from buying a newer card is lower power draw, but nothing would run any better because it’s already running at it’s maximum and doing it in an optimal way.
A screenshot below from various sensor readings during my testrun, this was for about half an hour but I have been playing Skyrim for 250+ hours at these settings without any problems. I have highlighted the constant fps in red at the bottom of the screenshot, see for yourself!
The card was running at 797MHz core, 1596MHz shaders and 3800MHz memory using an i7 2600K @ 4.8GHz. It’s of some interest that the cards framebuffer is almost full, slightly below 1280MB is the max reading, so a plain 1GB card would have suffered fps-dips.
Here’s the run by some guy with an i7 920:
http://www.sweclockers.com/forum/125-geforce/1098840-vilka-kommer-att-kopa-gtx-680-a-post12090450/
Compared to my old GTX470, moderately OC:ed at 738/3800, using a 2600K @ 4.8GHz:
The system with GTX680 has almost twice as fast graphics, but almost twice as slow physics. It’s not like my old card can’t run these tests so I guess I’m in no panic to upgrade, despite the obvious age of GTX470(I bought the absolutely first revision that came out). Real games aren’t as taxing on the system as 3DMark11.
If anything, I better wait for “Big Kepler” aka GTX685/695/780 to be able to see an even more dramatic difference in performance, even though upcoming GTX680 cards with custom board designs, custom coolers , better OC headroom and 4GB of RAM will be though to resist. It will even be though to resist a large pricecut on the AMD HD7970 when it occurs, but the same thing applies there about custom boards. Heck, my old GTX470 is definetely a reference design as I was a very early adopter back then. I will try not to repeat that, if nothing else, custom coolers are more quiet! 
Since Kubuntu isn’t officially supported by Canonical from now on, I thought it was time to once again try the Ubuntu flavour of their distro. I have really hated the Unity desktop so far, but it could be interesting to see how far it has developed. So, said and done. Now this server/workstation, which my site happens to reside on, is running yesterdays daily build of Ubuntu 12.04 Precise Pangolin. With all the latest updates. At least everything works, but it remains to be seen how well I will get along with Unity.
https://wiki.ubuntu.com/PrecisePangolin/TechnicalOverview/Beta1
Recent Comments