ViDock 2 - My experiences so far

[ViciousXUSMC]

Originally Posted by LaptopNut

I think in time the Vidock will be able to drive the laptops internal screen. If it never does this then I think they are shooting themselves in the foot. They must realise that being able to drive the internal laptops screen is essential for some of the market.

If this never happens then I will simply keep on buying gaming laptops with built in high end GPU's. There is no way in hell I am going to be limited to having to use a big external screen. I want to be able to game where ever I wish, not be limited by the technology, I want to be freed by it.

Yeah I have dual 4850's in the desktop and dual 4870's in the laptop

Its nice to be able to game on the go.

[MexicanSnake]

Nice video, good explanation it looks really good.

[nando4]

Request for performance data

The ViDock would be operating at whatever your expressport speed is (expressport 1.0), which I assume is 1x speed. The HD4670 capable of 16x speed in a 16x slot. So can anyone with a ViDock provide 3dmark06 results when using the 1x slot? notebookcheck gives 3dmark06 of HD4670 of 6995 in a 16x slot.

[masterchef341]

first off, the expressport is less than 1x pci-e 2.0

second, 3dmark gives "normal" results even with 1x speed (3dmark stresses the actual gpu as much as it can and therefore keeps everything in the GPU buffer)

games do not give normal results with 1x speed, depending on the card (i know for a fact that the 3870 and 9800gx2 lose 25-75% of game performance depending on the game at 1x, but lose basically nothing in 3dmark at 1x)

i 2nd the request for some performance data... i'm still really interested to see this. i just request that you don't use 3dmark. i *hope* that we will find that since we are using a 4600 series card instead of a 9800gx2, we won't see much drop in performance. still want to see the numbers to back this up.

[Steiner32]

Originally Posted by masterchef341

first off, the expressport is less than 1x pci-e 2.0

second, 3dmark gives "normal" results even with 1x speed (3dmark stresses the actual gpu as much as it can and therefore keeps everything in the GPU buffer)

games do not give normal results with 1x speed, depending on the card (i know for a fact that the 3870 and 9800gx2 lose 25-75% of game performance depending on the game at 1x, but lose basically nothing in 3dmark at 1x)

i 2nd the request for some performance data... i'm still really interested to see this. i just request that you don't use 3dmark. i *hope* that we will find that since we are using a 4600 series card instead of a 9800gx2, we won't see much drop in performance. still want to see the numbers to back this up.

I am just going to ask instead of guessing.... What do you want to see? I ran 3DMark06, Vantage and played with Crysis a little bit as well.

[masterchef341]

ok. i want numbers or graphs.

Crysis is good. the 9800gx2 at 1x saw about a 20% performance cut just rigged up normally in a desktop. I could imagine that means the 4600 series card might have *no performance drop* in Crysis? thats what i *hope*, but I just want to see the benchmark results.

So maybe take a few of the big names: Crysis, CoD4, Fallout 3, or whatever other intensive games you have available.

CoD4, in particular, is a good example of a game where the 9800gx2 suffered a lot at 1x. Also if you have flight simular x, that is a good benchmark. If these benchmarks turn out pretty good, that would be a good sign, because these two in particular really struggled at 1x with a 9800gx2.

[Steiner32]

I have COD4 installed on this laptop -- can try to figure out a way to test it (it is installed to an SSD in the expresscard slot) on Monday when I get into the office.

I ran Crysis tool (results were posted in the first thread) and here are the results.
Run #1- DX10 800x600 AA=No AA, 64 bit test, Quality: High ~~ Last Average FPS: 19.18
Run #2- DX10 800x600 AA=2x, 64 bit test, Quality: High ~~ Last Average FPS: 17.45
Run #3- DX10 800x600 AA=8xQ, 64 bit test, Quality: High ~~ Last Average FPS: 17.44
Run #4- DX10 1024x768 AA=No AA, 64 bit test, Quality: High ~~ Last Average FPS: 17.20
Run #5- DX10 1024x768 AA=2x, 64 bit test, Quality: High ~~ Last Average FPS: 15.51
Run #6- DX10 1280x960 AA=No AA, 64 bit test, Quality: High ~~ Last Average FPS: 14.57

If you would like I can try and create something pretty with that - but I am not a professional reviewer and don't even claim to be one on TV.

If you want to see numbers and test it out, why not just do what I did? Buy one.

[masterchef341]

Originally Posted by Steiner32

If you want to see numbers and test it out, why not just do what I did? Buy one.

I want to see the numbers before I test it out.

But yes, this crysis benchmark is spot on. I want to compare this performance to a similar laptop with a mobility 4600 series card with the same GPU clocks and shaders and such... might be a mobility 4650 or 4670... I'll try and find that stuff out.

thank you for the effort. +rep.

[SteelersWorship]

Tag and looking forward to the release of the power model.

[Aule]

Sure, it might be expensive, and I might be able to buy a cheap PC for the price, but I don't want a second computer. Say you've got a latest gen laptop, with C2D 2.8GHz, SSD drive and 6GB of RAM, but with a graphic card that dosn't really do games. Plug this in, and you have a gaming machine. This seems to be a better solution than buying a second computer solely to play games. (I really, REALLY don't understand why NO ONE ever offers a docking station with a proper PCI-Express, 16x; it's ridiculously simple, from an engineering point of view, to add a docking connector capable of doing this. Think about it, if you can have USB/Firewire/DVI/whatnot on a docking connector, why not a proper PCI-E???)

That said, I'd like to see cheaper version ;-)

What's the limiter on ViDock2? Why is 4670 the best card you can use? Is it physical size of the card or power supply? It shouldn't be too hard to just make a slightly bigger box!