Screen Vs Output and Affects on 3D Scores (The Experiment)

So A couple weeks ago I started asking around about how video outputs + screen size affect 3D Benchmark scores. I received a few mixed reactions to how these things affect scores. Needless to say I was not happy about the responses I got back . I felt it needed further investigation. So I grabbed some monitors and screens and set out to see how they would affect my scores.

THE SETUP: I was originally going to conduct these tests on my sig rig but my SLI setup just wasn't ticking properly, So I have employed my office rig for the job. (The Gears. See signature.) I will be running3Dmark01.

The testing will be done on a non optimized Win7 PC with no CPU or GPU overclocks.Running 3Dmark01 x3 times each screen. Will be testing across 5 different screen setups from CRT - 42".

THE SCREENS:
Ancient CRT
Asus 23.5" LED
Asus 23.5" LED x2 (Multiscreen.)
Panasonic Vierra 42" TV
Turned off screen (Asus 23.5" Just to see the effect.)

---

Screen>VGA Cable>VGA-DVI-D Adapter>5770.
Ok so background on this thing . It's an ancient Compaq screen the main reason I put the CRT in this experiment was to satisfy my own curiosity. Might also add some interesting scores.



Run 1


Run 2


Run 3


So the average here is 40084.86666 etc. Pretty low . Lets see how the 23.5" goes.

---

Screen>DVI-D Cable>5770.
Ok so this is my standard Office setup but usually in dual monitor. I will test this thing with every adapter combination I have available .Point where 300 above the CRT. Im doing the HDMI now.



Run 1


Run 2


Run 3


Averages Here are 40320.3333 etc.. So slightly up from the CRT.

---

Screen>HDMI Cable>5770.



Run 1


Run 2


Run 3


Averages Here are 40326 etc. So pretty much the same as DVI-D

---

Screen>VGA>VGA - DVI-D Adapter>5770.



Run 1


Run 2


Run 3


So the VGA cable with adapter kind of surprised me was expecting lower results .Averages Are 40279 .

---

Screen>HD Cable>5770Screen>HD Cable>5770.SCREEN TURNED OFF
So just in the name of equal testing let's try it with the screen turned off through a benchmark.



Run 1


Run 2
40218

Run 3
40197

So I only took screenies off the 1st Run I completely spaced on the other 2 Averages for screen turned off where 40210.33 etc. So nothing special on this front .
Edited by pwnography6 - 10/1/12 at 1:37am

Screen>HD Cable>DVI-D Adapter>5770.



Run 1


Run 2


Run 3


So the average here is 40210.3333 etc. Which is where I was expecting up next is the 42" T.V this is where I'm really expecting some differences in scores.

---

Screen>HDMI Cable>5770.

OK so First runs with the Big screen. This is my everyday screen and just wanted to see if it is affecting my regular overclocks.



Run 1


Run 2


Run 3


Averages Here are 40289.6666 etc.Which is a LOT higher than I was expecting really thought the screen size here would affect the scores. I have noticed you get a lot more tears and artifacts on the screen during the Tests but it just doesn't really seem to change the final result

---

Screen>VGA Cable>VGA - DVI-D Adapter>5770.



Run 1


Run 2


Run 3


Averages Are 40422 . Which once again are higher than I was expecting.

---

Screen>HD Cable>5770

So just to be even I thought I would run some Multi monitor tests. Screens are NOT identical but close enough to being for the sake of the experiment.



Run 1


Run 2


Run 3


Averages here are 40260.6666 etc. I don't normally bench multi monitors but this seemed like a decent score.
Edited by pwnography6 - 10/1/12 at 1:39am

Ok so put together a quick google spreadsheet of results for easier reading.It's my first google DOC so please go easy.



From the looks of my results it seems like unless you are using a really ancient CRT you don't get too much difference in scores. I will be continuing this experiment over the next few days as I am starting to wonder about a more GPU intensive test Like Heaven 11 . Would be Interesting to see if there is any change in points on the Direct X11 side of things. Hope this lil screen test gave someone a insight into the affects or lack of on Direct X9 .

Edited by pwnography6 - 10/1/12 at 1:40am

Reserved for Heaven 11

Edited by pwnography6 - 2/7/12 at 7:42am

Reserved for Heaven 11

Edited by pwnography6 - 2/7/12 at 7:42am

Yeah it was an interesting to see some results . Thanks for the support guys I have bulk running round to do today so I will try get the 1st lot of results out tonight will probably just be CRT and 23.5" Tests today as I wont have time to rob my main rig of it's screen.

The "point" of it all is to prove pretty much exactly that. I asked this question a billion times in different places and everyone told me it makes a massive difference. You are actually the 1st person to bring this up. Still if it signal and screen has no affect why do people put so much time and effort into what cable setup they use .

On top off that signal still does not explain why CRT screen results are 400 points lower. Im doing a public service here if people answer questions wrong and give misleading advice here on our OCN I would like to be able to give back to the forum and put some truth behind it all.

So what if majority of the results are similar iv still answered the question for someone else with proof behind my claims. It really seems like people dont care about testing unless there is massive variations in results but in my opinion this is equally important.

Post 600
Edited by pwnography6 - 2/7/12 at 7:20pm