Hardware & Technical Users System Specs

Greeboski

G
purely out of interest i ran it on this xp machine , as it would let me , but only in d3d9 (it was set to dx11 but auto swaped itself)

Unigine Heaven Benchmark 4.0
FPS: 43.4
Score: 1093
Min FPS: 10.7
Max FPS: 77.0
SystemPlatform: Windows XP (build 2600, Service Pack 3) 32bit
CPU model:AMD Athlon(tm) 64 Processor 3700+ (2210MHz) x1
GPU model:NVIDIA GeForce GTX 660 6.14.13.0527 (2048MB) x1
SettingsRender: Direct3D9
Mode:1280x720 fullscreen
Preset Custom
Quality Low
Tessellation: Disabled

Guess i dont win the rat race :p

will give another reading when i get my other sys , tho will only be one gtx 780 in (for now) after considering comments/ED's non-sli compliancy
 

Pyros

P
Liqua said:
I have a suggestion for you, but to be truely effective we need FD / Michaels help on this.

Listing components in a spreadsheet is a great start but with so many variants out there - manufacturers; basic clock speeds; drivers and settings - it can be difficult to make an informed decision.

What we need is a standard to judge against ... cue Heaven (3DMark you need to pay for)

It's free, tests DX11, takes up approx 250Mb of HDD space and the test lasts 260 seconds.

If we all run this test and record the results, including Michael on his minimum spec machine, then you have a like for like test to compare against.

Download the software, run it, and set the following:
  • Preset: Custom
  • API: DirectX11 (as FD are only supporting this)
  • Quality: Low (to give low end users a chance)
  • Tessellation / 3D / Multi-monitor / Anti-aliasing: off
  • Full screen: ticked
  • Resolution : 1280x800 (closest one to the minimum Michael said)

Hit RUN .. click the sound button at the top to silence it, then press benchmark. 260 seconds later you will get some results:

My laptop, for instance:
  • FPS: 108.3
  • Score: 2728
  • Min FPS: 9.4
  • Max FPS: 172.5

This should give your spreadsheet a standardised result column for comparison purposes.
Click to expand...

Excellent idea. Also adding the standard test config to the spreadsheet can be useful.
My results:
  • FPS: 228.3
  • Score: 5751
  • Min FPS: 9.7
  • Max FPS: 386.5

Settings:
  • Render: Direct3D11
  • Mode: 1280x800 fullscreen
  • Preset Custom
  • Quality Low
  • Tessellation: Disabled
 
Last edited:

Greeboski

G
Pyros said:
8 gb Ram minimum?:S where did you see that?
Click to expand...

he never. the info we were given was . . .

Default Minimum Specs for Alpha Test
I know there's been some discussion about the machine specs needed for the alpha test, as we're approaching the first build delivery I can share some of the development specs of where we are at the moment. Optimisation and performance are obviously ongoing tasks and as more features are added through the alpha and beta periods we'll need to keep revisiting it.

To play the first build released in December we're currently looking at the following:

Direct X 11
Quad Core CPU ( 4 x 2Ghz is a reasonable minimum)
2 GB System Ram (more is always better)
DX 10 hardware GPU (my reasonably low end machine has a GTX 285 with 2GB)

As I say we're looking to try and reduce the spec if possible through detail options and optimisations, so the final min spec for the game should be lower than that stated above, but for the initial alpha tests you'll need to be in that range.

Michael
 

Asp Explorer

Asp Explorer
I wonder if anyone will add an "estimated cost" column to the spreadsheet - see just how much money we've invested in hardware to throw at testing ;)

I had a good laugh upon looking at some of the benchmarks.

Manufacturer / Model / Speed / Size / Latency /Read / Write / Price
Corsair CMZ32GX3M4X1866C10 8GB 25 13661 10395 $379.99
G Skill Intl F3-12800CL9-4GBXL 4GB 24 14136 9573 $40.99

I'd also like to propose a "Links, Pics, or it never happened" system - if you are going to post numbers, they should have some verifiable evidence to back them up.
 
Last edited:

Aewarin Jameson

Aewarin Jameson
Pyros said:
Excellent idea. Also adding the standard test config to the spreadsheet can be useful.
My results:
  • FPS: 228.3
  • Score: 5751
  • Min FPS: 9.7
  • Max FPS: 386.5

Settings:
  • Render: Direct3D11
  • Mode: 1280x800 fullscreen
  • Preset Custom
  • Quality Low
  • Tessellation: Disabled
Click to expand...

Very nice Pyros but I would suggest you redo this using 3DMark11 - it seemed someone else had a similar thought and whilst I was making my post they added the column to the spreadsheet and that software is more widely known / standardised :)

Asp Explorer said:
I'd also like to propose a "Links, Pics, or it never happened" system - if you are going to post numbers, they should have some verifiable evidence to back them up.
Click to expand...
I think this is a little OTT .. does it really matter if someone has a bigger score than you ? :eek:
 

Styggron

S
Steph said:
Stop it! You're scaring me!

I've only just taken delivery of my laptop (space is at a premium here). I got the best graphics card that was offered with it (NVidia GTX 780M) - if that isn't enough then I don't know what is....
Click to expand...

LOL, well the work laptop had the 750M and those tests were barely getting 10FPS.......

Don't worry. my laptop got a score of 763 and this is what I am going to run ED on so you'll be right..as for me, graphics on lowest setting and if there is wireframe I'll go for that too !
 

Zieman

Zieman
Steph said:
Stop it! You're scaring me!

I've only just taken delivery of my laptop (space is at a premium here). I got the best graphics card that was offered with it (NVidia GTX 780M) - if that isn't enough then I don't know what is....
Click to expand...

Styggron said:
LOL, well the work laptop had the 750M and those tests were barely getting 10FPS.......

Don't worry. my laptop got a score of 763 and this is what I am going to run ED on so you'll be right..as for me, graphics on lowest setting and if there is wireframe I'll go for that too !
Click to expand...
You really have no worry Steph, one GTX 780M is worth 2 750Ms in SLI or more.
game-debate.com said:
In terms of overall gaming performance, the graphical capabilities of the Nvidia GeForce GTX 780M are massively better than the Nvidia GeForce GT 750M SLI
Click to expand...
 

Sir.Tj

The Moderator who shall not be Blamed....
Volunteer Moderator
Sir.Tj
My poor old PC couldn't even run the Heaven program without crashing.

Beat that. :D
 

Pyros

P
Asp Explorer said:
I wonder if anyone will add an "estimated cost" column to the spreadsheet - see just how much money we've invested in hardware to throw at testing ;)

I had a good laugh upon looking at some of the benchmarks.

Manufacturer / Model / Speed / Size / Latency /Read / Write / Price
Corsair CMZ32GX3M4X1866C10 8GB 25 13661 10395 $379.99
G Skill Intl F3-12800CL9-4GBXL 4GB 24 14136 9573 $40.99

I'd also like to propose a "Links, Pics, or it never happened" system - if you are going to post numbers, they should have some verifiable evidence to back them up.
Click to expand...

well, regarding prices, while interesting in principle, not only there is the problem that HW tend to vary and sometimes quite suddenly (a GTX780 bought a month ago was £80+ more expensive than today), as also the prices vary a lot from country to country (taxes, market size,...). So the info that would take some work to post would be of little to no value to may others.

Regarding proof, well, this is not a "my HW is better than yours" or a OC record forum, just a guide for other fellow player to know what to expect from different HW configs. Best course of action if you see some strange score is to ask the person. He might have made a mistake or misinterpreted something.
 

Aewarin Jameson

Aewarin Jameson
Steph said:
Stop it! You're scaring me!

I've only just taken delivery of my laptop (space is at a premium here). I got the best graphics card that was offered with it (NVidia GTX 780M) - if that isn't enough then I don't know what is....
Click to expand...

Zieman is right - you have nothing to worry about.

Again, not suggesting this site is perfect but useful as a guide:

  • GTX285 (reference card) : 1254
  • GT750M : 1557
  • GTX780M : 4247

Styggrons laptop should be OK for ED in low settings - I suspect yours will work with considerably higher ones.
 

bitrogue

bitrogue
Liqua said:
  • GTX780M : 4247
Click to expand...

Taking delivery of mine next week. But when I saw that the speed of the real GTX780 was almost double that of the mobile version I started having serious doubts about my choice. Still, that chart says its ranked 22nd so that must count for something. I hope. Maybe. :rolleyes:
 

Geraldine

Geraldine
Moved thread to Hardware & Technical forum section. Great idea about the spreadsheet. Those system specs could prove very useful to FD :cool:
 

Rog

Rog
Just added a (dreadful) Passmark score (979.6).

Settings: so do we just download, run it and hit "run benchmark" with no alterations, and then take the overall score?

Or is it the 3d graphics score that's required?
 

Asp Explorer

Asp Explorer
Pyros said:
Regarding proof, well, this is not a "my HW is better than yours" or a OC record forum, just a guide for other fellow player to know what to expect from different HW configs.
Click to expand...

Oh, I quite agree. No wanger-waving was intended, merely curiosity. My hardware is average at best, many people have scores much greater than mine, and I don't overclock as it reduces stability. The reason I mentioned it is because some of these benchmarks are simply wrong.

My RAM is detected and listed by some of these programs at 667MHz. Others at 800MHz. GPU and PC have 0 RAM available to them in some. There is simply no consistency, even between benchmarks on the same hardware.

I work on big iron - and that simply would not stand in the mainframe world. If someone has paid X amount of $ for Y amount of performance, and they don't get it - absolutely all hell breaks loose. It can be very awkward for all parties.

It's just simply staggering that people will put such weight behind benchmarks, and indeed buy hardware on their results, yet those very benchmarks get so many things wrong.
 
Last edited:

rab

R
They can report the wrong speeds at times, sometimes your CPU and GPU will get detected at there idle or 2d clock speeds in stead of full load speeds, and I think with the RAM it can get detected at the rated speed of the ram instead of it's actual speed it is running at.

Also as Rog (love the name btw, Trapdoor fan?)asked what are people doing for the Passmark test running the full test or just the GPU test?
 
You must log in or register to reply here.
Back
Top Bottom