Returned Alienware

Sorry to see you go Sajin. BTW what did the Alienware do with 3 gtx 480's? Did you ever get a chance to try it before you sent it back?

You cannot run 3x GTX 480's in any alienware the PSU is to weak, you would have to change the PSU losing all lighting & vents to run 3x GTX 480 or 3x anything for that matter. Also the 3rd PCI-E slot on all alienware motherboards run at 8x you cannot change it so you would be losing performance with the 3rd card not running at its full potential which should be 16x like the two other PCI-E slots. When I called alienware tech support they told me the 3rd PCI-E slot was suppose to be used with low power cards such as sound cards, wireless cards, etc.

I ran 3 gtx 285's just fine in mine and now im running 2 gtx 295's and one gtx 285. The 1100watt psu can handle it. If not you can always use a PSU booster. I have two of them that run at 450watt and peak at 500watt. I used them on my old 730 and 730x systems that only had a 1k PSU for 3-way sli. They fit in the CD drive bay and cost about $70 bucks each.   Below are two picks of the XPS 730X with the 3 GTX 285's and one 500watt Booster X5 in the top bay and two pics of my Area 51 with not only 5gpu's in it but 2 extra 120mm fans up top,one 92mm rear fan and the 2 Corsair GT fans. Man I miss my XPS 730x now. It was one pretty machine and a true beast at the time. Again my area 51 was running 3-way sli and is now running 5 gpu's and 5 extra fans all of that on the 1100 PSU.

 http://www.newegg.com/Product/Product.aspx?Item=N82E16817104054

As you can see my GTX 295's ( and you know the cards well ) are pulling more then the GTX 480's. Running them OC and with the 285 for PhysX has yet to shut me down and the day it does I will reach in my closet and whip out one of my FSP Booster X5's:) Now I know the gtx 285 aint doing much but even runnning PhysX benchmarks it has never shut me off. When I was running the 3-way SLI with the 285's it still didnt shut down and I ran benchmarks up the hoohoo with it overclocked to the max. Now when I had the XPS systems with the 1k I did get shut downs but the booster solved that.

As for the PCI-E lanes its true that when you use all 3 slots on the Area-51 they run at x16,x16,x8.. Here is a post I made a while back when i had my XPS 730X system. Go thru it. Click the links. Google it. You will see that running x8 is still more then enough for everything but the GTX 480 and even then its very close and you can not notice it. Also that 4-way sli board will not give you anymore performance. The N200 chips are garbage and are faking x16,x16,x16. You cant get something from nothing. So running 3 GTX 480's on your EVGA board or the alien boards makes no difference. Your chip is pausing one lane to send the other thru at x16. It can not and never will be able to send all 3 lanes thru at 16x at the same time because the i7 x58 chipset doesnt have that many lanes. The x58 has 36 lanes while true x16,x16,x16 would require 48 lanes. This pausing( or shared bandwidth ) kills whatever performance you may have gotten from the extra x8. Other then 3Dmark benchmarking you are not getting any real world perfromance from it.

Also in this review the GTX 480 running at PCI 2.0 x8 in 2 slots there is a 2% drop on some games. Yeah a whole 2%. And that was with two slots running at x8. On the Alienware you would only have one running at x8 ( with 3-way ) so you would see maybe a 1% drop on some games when fully stressed.

Here is the last page conclusion from the first link below. Keep in mind this is with two x8 vs two x16 boards. Area-51 would run 2 of the 3 at x16 and only the last at x8 like mine is running now which would be even less if any of a perfromance hit. If you got time read the entire review.

** NVIDIA's GeForce GTX 480 is a very fast graphics processor. To maintain its speeds, it would hypothetically require high system bandwidth, leading one to think that lesser PCI-Express configurations would cripple it. The theory couldn't be more wrong, as seen by the mere 2% performance loss going from x16 to x8 (which reduces bandwidth by 50%). To cite results from one of the latest and resource-heavy games in our bench, Collin McRae DiRT 2, that translates into something like 63.2 FPS vs. 62.1 FPS, at 2560 x 1600 pixels resolution - barely a difference.
We also examined how much PCI-E 2.0 x4 (or PCI-E 1.1 x8, older motherboards) would affect the GTX 480. And the result coarsely put is "not much": 8%. The other part of the analysis is to see how the GTX 480 scales with decreases in interconnect bandwidth compared to AMD's Radeon HD 5870. When compared to the card from the red camp, the GeForce GTX 480 takes bigger hits with decrease in bandwidth. It is scaling worse than the HD 5870, which could be because the GPU is faster and makes better use of available bandwidth, and hence gets the pinch due to lack of it. That still shouldn't come in the way of you using two GTX 480 cards in SLI on a motherboard with electrical x8 PCI-E slots. You shouldn't be worried about using these cards on a PCI-E 1.1 motherboard, either.**

http://www.techpowerup.com/reviews/NVIDIA/GTX_480_PCI-Express_Scaling

http://www.engadget.com/2010/04/26/nvidia-geforce-gtx-480-4-way-sli-exemplifies-law-of-diminishing/

http://www.tomshardware.com/reviews/eclipse-plus-n200,2332.html

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

**This is an old post I made way back when on the XPS730 boards but I figured I'd post it here to since it covers the same topic.**

---------------------------------------------------------

I had this posted on the old thread so I will just copy it here so you can see what the speeds are of the lanes and see that 3 PCI-E slots running at X16 each is only done with the Nvidia 200 chip and it slows the system down not speed it up because it must pause a lane to send on another at x16.  In other words 2 lanes are shared. Remember that the i7 northbridge does not have the 48 lanes which would be needed. Click the link and read the review.  

  The XPS 730x Mobo PCI-E slots will run at x16,x16,x4 with Dual SLI and x16,x8,x8 with Tri SLI(order from bottom to top on Mobo). The bandwidth on the PCI-E 2.0 at x8 is 4gb/s. A GTX 280 only uses 2.1GB/s, 2.4 GB/s for the GTX 285 and 3.9GB/s for the GTX 295. Go download GPU-Z and you can see the lanes being used.

 Here is a review of a true X16,x16,x16 Mobo VS one like ours with X16,x8,x8. Read it before you go buying a new Mobo.http://www.tomshardware.com/reviews/eclipse-plus-n200,2332.html 

Also check this out. This helps explain it:

___________________________________________________________________________________________________________________________________________

It seems many threads all over the internet has/had/have asked this same question all over.

was tweaktown's review really truthful ? sadly.. it wasn't.

Difference between 16x PCI-e vs 8x PCI-e in CrossFire/SLI ?

Answer: None

Here is why.

PCI-e 2.0 16X = 8GB/s total bandwidth
PCI-e 2.0 8X = 4GB/s total bandwidth

ATI:

HD 3870 = 1125mhz (dual piped - 2.2 GB/s data rate) = 2.2 GB/s total data rate
HD 4650 = 500mhz (dual piped - 1 GB/s data rate) = 1 GB/s total data rate
HD 4670 = 1000mhz (single piped - 2 GB/s data rate) = 2 GB/s total data rate
HD 4770 = 800mhz (quad piped - 3.6 GB/s data rate) = 3.6 GB/s total data rate
HD 4830 = 900mhz = (dual piped - 1.9 GB/s data rate) = 1.9 GB/s total data rate
HD 4850 = 993mhz (dual piped - 1.9 GB/s data rate) = 1.9 GB/s total data rate
HD 4850 X2 = 993mhz (dual piped - 1.9 GB/s data rate)(times 2) = 3.8 GB/s total data rate
HD 4870 = 900mhz (quad piped - 3.6 GB/s data rate) = 3.6 GB/s total data rate
HD 4890 = 975mhz (quad piped - 3.9 GB/s data rate) = 3.9 GB/s total data rate
HD 4870 X2 = 900mhz (quad piped - 3.6 GB/s data rate)(times 2) = 7.2 GB/s total data rate

So as you can see, the 4870 X2 will be hurt in an 8X condition and why the 3870, 4650, 4670, 4770, 4830, 4850, 4870, 4890, and 4850 X2 will not be hurt in performance at all.

Nvidia:

8800 Ultra = 1080mhz (dual piped - 2.1 GB/s data rate) = 2.1 GB/s data rate
9600 GT = 900mhz (dual piped - 1.8 GB/s data rate) = 1.8 GB/s total data rate
9800 GTX = 1100mhz (dual piped - 2.2 GB/s data rate) = 2.2 GB/s total data rate
GTS 250 = 1100mhz (dual piped - 2.2 GB/s data rate) = 2.2 GB/s total data rate
GTX 260 216 = 999mhz (dual piped - 1.9 GB/s data rate) = 1.9 GB/s total data rate
GTX 275 = 1134mhz (dual piped - 2.2 GB/s data rate) = 2.2 GB/s total data rate
GTX 280 = 1053mhz (dual piped - 2.1 GB/s data rate) = 2.1 GB/s data rate
GTX 285 = 1242mhz (dual piped - 2.4 GB/s data rate) = 2.4 GB/s data rate
GTX 295 = 999mhz (dual piped - 1.9 GB/s data rate)(times 2) = 3.8 GB/s total data rate
GTX 285X2 = 1242mhz (dual piped - 2.4 GB/s data rate)(times 2) = 4.8 GB/s data rate

So as you can see, the GTX285 X2 will be hurt in an 8X condition and why the 8800, 9600, 9800, 250, 260, 280, 275, 285 & 295 will not be hurt in performance at all.


This should be a clear message to anyone wanting to do multi-gpu setup's, people with PCI-e 2.0 @ 16x-16x-8x or 16x-8x-8x or even 8x-8x-8x bandwidth should have no bottle neck for any single gpu cards in the market, aside from the double gpu cards like 4870x2, and incoming GTX285x2.

a quick example:

Which is better performance in fps/gaming, 16x-8x-8x w/ PCI-e 2.0 motherboard vs 16x-16x-16x w/ PCI-e 2.0 motherboard vs 8x-8x-8x w/ PCI-e 2.0 motherboard:
Code:

1x 4870 1GB @ 16x slot
1x 4870 1GB @ 8x slot
1x 4870 1GB @ 8x slot
vs
1x 4870 1GB @ 16x slot
1x 4870 1GB @ 16x slot
1x 4870 1GB @ 16x slot
vs
1x 4870 1GB @ 8x slot
1x 4870 1GB @ 8x slot
1x 4870 1GB @ 8x slot

Winner:
Draw, no difference in fps. Although some 16x-16x-16x boards with n200 chipset's has fps drops in games which is reviewed many times but no drops in benchmarking (3dmarks,vantage). If your board is PCI-e 1.0 then that would be a huge difference

Ok so I tried to max my cards to get it to crash. I ran furmark first. This would only run the 295's. I then tried Fluidmark. This is a physX benchmark so it got my gtx 285 going but it would only run one gpu on one of my gtx 295's. So I tried Dark Void... meh. The benchmark sucks. While it would run all four of the GPU's on the 295's and do the PhysX on the 285 I couldnt stress it enough.

So I said the heck with it and broke out my Batman Arkham Asylum. After half an hour to install and update it I ran my first benchmark. It got my 295's up tp 65% and at one point my 285 hit 59%. With no crashing I went and bumped my Antialiasing from enhance the app 2x to override any app SLI 64xQ. I then turned on Ambient Occusion on full along with antialiasing-transparency to supersampling. And last I put Texture filtering on high quality. Every setting was put to absolute max.

Ran it again and BAM! I got a restart about 5secs into the benchmark. Tried it 3 times like this and all 3 times gave me a restart. So now i'v hit the power limit of the 1100watt PSU. So like I said before if it ever happened I would open my closet and bust out one of my FSP boosters and thats what I did. Took me about 15min to install it. I am now effectively running 1600watts.

Went back to Batman AA and ran the benchmark again with every setting on max. All 4 GTX 295 GPU's ran in the 95% range going up and down by 5%. My GTX 285 hit 73%. And people say anything over a 9800 is overkill for PhysX..pffft. So anyway I got a low 59fps/high 179fps/avg 123fps.

Here is a pick of my FPS booster in my Area 51. Like i said i have 2 of these so if someday comes I need to add another I could pull one of my Blu-ray burners and put it in but I dont think I'll ever need that. But then again Im holding out for the GTX 495 dual GPU cards so who knows. Do wish these comps had 4 drive bays like the XPS systems did. Im starting to miss that extra bay. You can change to color of the X to match whatever color you have AlienFX set to.

Try running 3DMark Vantage on performance preset or try running OCCT's Power Supply Test.

I ran 3 gtx 285's just fine in mine and now im running 2 gtx 295's and one gtx 285. The 1100watt psu can handle it. If not you can always use a PSU booster. I have two of them that run at 500watt. I used them on my old 730 and 730x systems that only had a 1k PSU for 3-way sli. They fit in the CD drive bay and cost about 60 bucks each.   Below are two picks of the XPS 730X with the 3 GTX 285's and one 500watt Booster X5 in the top bay and two pics of my Area 51 with not only 5gpu's in it but 2 extra 120mm fans up top,one 92mm rear fan and the 2 Corsair GT fans. Man I miss my XPS 730x now. It was one pretty machine and a true beast at the time. Again my area 51 was running 3-way sli and is now running 5 gpu's and 5 extra fans all of that on the 1100 PSU.    As you can see my GTX 295's ( and you know the cards well ) are pulling more then the GTX 480's. Running them OC and with the 285 for PhysX has yet to shut me down and the day it does I will reach in my closet and whip out one of my FSP Booster X5's:) Now I know the gtx 285 aint doing much but even runnning PhysX benchmarks it has never shut me off. When I was running the 3-way SLI with the 285's it still didnt shut down and I ran benchmarks up the hoohoo with it overclocked to the max. Now when I had the XPS systems with the 1k I did get shut downs but the booster solved that.

Your telling me that your Area-51 with it's 1100 watt psu from dell can support 1575 watt's when under load? But your old XPS system with it's 1000 watt psu shut down on you with 3x 285 which equals 1125 watts?  lol, I don't think so...

As for the PCI-E lanes its true that when you use all 3 slots on the Area-51 they run at x16,x16,x8.. Here is a post I made a while back when i had my XPS 730X system. Go thru it. Click the links. Google it. You will see that running x8 is still more then enough for everything but the GTX 480 and even then its very close and you can not notice it. Also that 4-way sli board will not give you anymore performance. The N200 chips are garbage and are faking x16,x16,x16. You cant get something from nothing. So running 3 GTX 480's on your EVGA board or the alien boards makes no difference. Your chip is pausing one lane to send the other thru at x16. It can not and never will be able to send all 3 lanes thru at 16x at the same time because the i7 x58 chipset doesnt have that many lanes. The x58 has 36 lanes while true x16,x16,x16 would require 48 lanes. This pausing( or shared bandwidth ) kills whatever performance you may have gotten from the extra x8. Other then 3Dmark benchmarking you are not getting any real world perfromance from it.   Also in this review the GTX 480 running at PCI 2.0 x8 in 2 slots there is a 2% drop on some games. Yeah a whole 2%. And that was with two slots running at x8. On the Alienware you would only have one running at x8 ( with 3-way ) so you would see maybe a 1% drop on some games when fully stressed.   Here is the last page conclusion from the first link below. Keep in mind this is with two x8 vs two x16 boards. Area-51 would run 2 of the 3 at x16 and only the last at x8 like mine is running now which would be even less if any of a perfromance hit. If you got time read the entire review.   ** NVIDIA's GeForce GTX 480 is a very fast graphics processor. To maintain its speeds, it would hypothetically require high system bandwidth, leading one to think that lesser PCI-Express configurations would cripple it. The theory couldn't be more wrong, as seen by the mere 2% performance loss going from x16 to x8 (which reduces bandwidth by 50%). To cite results from one of the latest and resource-heavy games in our bench, Collin McRae DiRT 2, that translates into something like 63.2 FPS vs. 62.1 FPS, at 2560 x 1600 pixels resolution - barely a difference. We also examined how much PCI-E 2.0 x4 (or PCI-E 1.1 x8, older motherboards) would affect the GTX 480. And the result coarsely put is "not much": 8%. The other part of the analysis is to see how the GTX 480 scales with decreases in interconnect bandwidth compared to AMD's Radeon HD 5870. When compared to the card from the red camp, the GeForce GTX 480 takes bigger hits with decrease in bandwidth. It is scaling worse than the HD 5870, which could be because the GPU is faster and makes better use of available bandwidth, and hence gets the pinch due to lack of it. That still shouldn't come in the way of you using two GTX 480 cards in SLI on a motherboard with electrical x8 PCI-E slots. You shouldn't be worried about using these cards on a PCI-E 1.1 motherboard, either.**   http://www.techpowerup.com/reviews/NVIDIA/GTX_480_PCI-Express_Scaling   http://www.engadget.com/2010/04/26/nvidia-geforce-gtx-480-4-way-sli-exemplifies-law-of-diminishing/   http://www.tomshardware.com/reviews/eclipse-plus-n200,2332.html       ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------   **This is an old post I made way back when on the XPS730 boards but I figured I'd post it here to since it covers the same topic.**   ---------------------------------------------------------   I had this posted on the old thread so I will just copy it here so you can see what the speeds are of the lanes and see that 3 PCI-E slots running at X16 each is only done with the Nvidia 200 chip and it slows the system down not speed it up because it must pause a lane to send on another at x16.  In other words 2 lanes are shared. Remember that the i7 northbridge does not have the 48 lanes which would be needed. Click the link and read the review.           The XPS 730x Mobo PCI-E slots will run at x16,x16,x4 with Dual SLI and x16,x8,x8 with Tri SLI(order from bottom to top on Mobo). The bandwidth on the PCI-E 2.0 at x8 is 4gb/s. A GTX 280 only uses 2.1GB/s, 2.4 GB/s for the GTX 285 and 3.9GB/s for the GTX 295. Go download GPU-Z and you can see the lanes being used.    Here is a review of a true X16,x16,x16 Mobo VS one like ours with X16,x8,x8. Read it before you go buying a new Mobo.http://www.tomshardware.com/reviews/eclipse-plus-n200,2332.html        Also check this out. This helps explain it:   ___________________________________________________________________________________________________________________________________________   It seems many threads all over the internet has/had/have asked this same question all over. was tweaktown's review really truthful ? sadly.. it wasn't. Difference between 16x PCI-e vs 8x PCI-e in CrossFire/SLI ? Answer: None Here is why. PCI-e 2.0 16X = 8GB/s total bandwidth PCI-e 2.0 8X = 4GB/s total bandwidth ATI: HD 3870 = 1125mhz (dual piped - 2.2 GB/s data rate) = 2.2 GB/s total data rate HD 4650 = 500mhz (dual piped - 1 GB/s data rate) = 1 GB/s total data rate HD 4670 = 1000mhz (single piped - 2 GB/s data rate) = 2 GB/s total data rate HD 4770 = 800mhz (quad piped - 3.6 GB/s data rate) = 3.6 GB/s total data rate HD 4830 = 900mhz = (dual piped - 1.9 GB/s data rate) = 1.9 GB/s total data rate HD 4850 = 993mhz (dual piped - 1.9 GB/s data rate) = 1.9 GB/s total data rate HD 4850 X2 = 993mhz (dual piped - 1.9 GB/s data rate)(times 2) = 3.8 GB/s total data rate HD 4870 = 900mhz (quad piped - 3.6 GB/s data rate) = 3.6 GB/s total data rate HD 4890 = 975mhz (quad piped - 3.9 GB/s data rate) = 3.9 GB/s total data rate HD 4870 X2 = 900mhz (quad piped - 3.6 GB/s data rate)(times 2) = 7.2 GB/s total data rate So as you can see, the 4870 X2 will be hurt in an 8X condition and why the 3870, 4650, 4670, 4770, 4830, 4850, 4870, 4890, and 4850 X2 will not be hurt in performance at all. Nvidia: 8800 Ultra = 1080mhz (dual piped - 2.1 GB/s data rate) = 2.1 GB/s data rate 9600 GT = 900mhz (dual piped - 1.8 GB/s data rate) = 1.8 GB/s total data rate 9800 GTX = 1100mhz (dual piped - 2.2 GB/s data rate) = 2.2 GB/s total data rate GTS 250 = 1100mhz (dual piped - 2.2 GB/s data rate) = 2.2 GB/s total data rate GTX 260 216 = 999mhz (dual piped - 1.9 GB/s data rate) = 1.9 GB/s total data rate GTX 275 = 1134mhz (dual piped - 2.2 GB/s data rate) = 2.2 GB/s total data rate GTX 280 = 1053mhz (dual piped - 2.1 GB/s data rate) = 2.1 GB/s data rate GTX 285 = 1242mhz (dual piped - 2.4 GB/s data rate) = 2.4 GB/s data rate GTX 295 = 999mhz (dual piped - 1.9 GB/s data rate)(times 2) = 3.8 GB/s total data rate GTX 285X2 = 1242mhz (dual piped - 2.4 GB/s data rate)(times 2) = 4.8 GB/s data rate So as you can see, the GTX285 X2 will be hurt in an 8X condition and why the 8800, 9600, 9800, 250, 260, 280, 275, 285 & 295 will not be hurt in performance at all. This should be a clear message to anyone wanting to do multi-gpu setup's, people with PCI-e 2.0 @ 16x-16x-8x or 16x-8x-8x or even 8x-8x-8x bandwidth should have no bottle neck for any single gpu cards in the market, aside from the double gpu cards like 4870x2, and incoming GTX285x2. a quick example: Which is better performance in fps/gaming, 16x-8x-8x w/ PCI-e 2.0 motherboard vs 16x-16x-16x w/ PCI-e 2.0 motherboard vs 8x-8x-8x w/ PCI-e 2.0 motherboard: Code: 1x 4870 1GB @ 16x slot 1x 4870 1GB @ 8x slot 1x 4870 1GB @ 8x slot vs 1x 4870 1GB @ 16x slot 1x 4870 1GB @ 16x slot 1x 4870 1GB @ 16x slot vs 1x 4870 1GB @ 8x slot 1x 4870 1GB @ 8x slot 1x 4870 1GB @ 8x slot Winner: Draw, no difference in fps. Although some 16x-16x-16x boards with n200 chipset's has fps drops in games which is reviewed many times but no drops in benchmarking (3dmarks,vantage). If your board is PCI-e 1.0 then that would be a huge difference

Seven Words: Take it like a grain of salt.

or to really test everything run the last level of crysis which is reckoning and fight the two bosses... if you have crysis.

Furmark burn is absolute max. You wont get anything near that for power usage in real life. It just doesnt happen. I figured you knew that since you know a lot about computers. You always gotta shave some off the final score with Furmark for a real life score. If you want RL for what 3 GTX 480s in 3-way SLI will do here ya go.

This is full system load with an intel 980x OC to 4.4ghz taken from http://translate.googleusercontent.com/translate_c?hl=en&ie=UTF-8&sl=nl&tl=en&u=http://www.hardware.info/nl-NL/articles/amdnampoZGCa/Clash_of_the_Titans_3way_SLI_GTX_480_test/15&prev=_t&rurl=translate.google.com&usg=ALkJrhi7eiZC34eoQHAmz9SzgU8Yno2gGQ

They ran all the tests on a 1200 watt PSU.

Yes I know. It was just very late last night and I was tired.

Yes I know. It was just very late last night and I was tired.

Post some pics of your new rig I would love to see it.

If one 5870 needs 289watts, how come 2 don't need 578 watts? (just an example).

Maybe Area51s are different, but I priced out an exact duplicate (as close as I could get it) of my Aurora-x58 at NewEgg before I bought it. Actually it was a little more. While both will handled dual video cards, I'm never planning to do that. If anything, I'll sell my old card and buy a single dual-gpu card.

Thanks for showing the FSP Group Booster products. I've never seen those before. So, how do you get both power supplies to come on at the same time (or isn't that important)?

Thanks for showing the FSP Group Booster products. I've never seen those before. So, how do you get both power supplies to come on at the same time (or isn't that important)?

You plug a 4pin molex cable into the back of booster. The white connector shown here. When the system comes on this sends a power signal to the Booster telling it to come on. My Area 51 has one 4pin molex coming off the sata power connector for the BR drive. The booster has its own power cord ( the black connector ) that goes out the back of the system and plugs into the wall and only uses the 4pin molex to tell it to turn on/off.

The booster has two PCI-e 6pin connectors and two PCI-e 6/8 pin connectors that plug into the video cards. They are the blue and red cables you see at the very back of the unit.

This is the actual power cord going out the back of the system. You will need a free slot on the back of your PC for it. Just remove the plate and put this in its place.

Let me get some pics.

Here you go...

Uploaded with ImageShack.us

Uploaded with ImageShack.us

Uploaded with ImageShack.us

Uploaded with ImageShack.us