If you had to pick one, which one would you pick? A8N32-SLI or Expert? Please don't tell me "Both are good", "They're different animals" or anything in that sense. Most of us have to pick ONE and that exact question is what we want to know. I'd venture to say what matters are following two:
1. Overclockability
2. Stability
Fetures, layouts, etc... yeah.. they are all good and nice, but what really matters are whether the overclock is stable. And that's what brought DFI here today. Could you comment on it? If you had a choice to pick JUST ONE, which one would you pick?
Great review anyway. I'm actually happy that AT is becoming more enthusiast-friendly and looking forward to the next review. (Possibly Opteron Overclocking review?)
I'll be honest in telling you I have not had my hands on the Asus board to know it's full overclocking abilities, so I really can't give you an opinion on which board to choose. Wesley could give you full insight on that one.
I go way back in the overclocking business, having hooked up with my best buds OPPAINTER and DDTUNG back in the day. At that time we were modding and overclocking each generation of Abit AMD based boards starting with the KG7 and culminating with the NF7-S, which we were pre-production testing prior to it's retail release. I've also had a play with Epox, Gigabyte, and the famous Shuttle A64 board.When I recieved the original SLI-D back in January it was an overclockers dream come true.All the voltage and overclocking/memory options and it overclocked way easy compared to the norm.(Thank you Oskar Wu:))
The Expert is more of the same but much better. I am a bit biased on a personal level about the board, and if you had just asked about it, I'd have given a thumbs up.
Under the extreme oc page, the pic showing the HTT @ 400MHz and RAM @ 300MHZ 2.5-3-3-8 but the sisoft is only 6674 MB/s? 300MHz should be displaying something near 8GB/s
That would be true if not running at 2400Mhz.All HTT and Memory overclocking tests were done at a reduced multiplier. Each multiplier also has an effect on system bandwidth in the way the A64 responds to them.If I had chosen to run 7 and a total CPU Mhz of 2800, you would have most likely seen the 8000mb/s figure.
In contrast, the stock 4000+ at 12x200Mhz with tight timings will average 5600-5700mb/s.
Another consideration is that with the bios used which made it more ram overclocking friendly, that there was possibly a relaxation of some of the bios register settings, which can also reduce bandwidth a bit in the name of pushing the memory higher in Mhz.
What i'm really interested in is if this board suffers from the same problem as my Ultra-D with PC4000 VX memory (the infamous cold boot issue)?
I don't really want to pick this board up and have it suffer from the same problem. Not that i've tried my VX with the latest beta bios, but that's besides the point.
I did not test with 4x 1GB sticks because I only have 2 in my possession for reviewing. I did in fact as mentioned run 4x512MB sticks at 2T with no problem whatsover for testing.
Future reviews will very well see x2 A64's included but I'm not sure what the timetable will be for their inclusion.
Thanks Randi for the very detailed review of the Expert.
I was wondering, with the new layout of the RAM slots and CPU, is it possible to populate all 4 RAM slots and still be able to fit the thermalright XP-120 without problems?
Bravo to DFI for using decent power supplies and such, but why is the SIL 3114 such a dissapointment? Does any desktop user need SATA2 at this time? Does anyone need DDR2 as opposed to high quality DDR (not this board but in a couple months). This whole idea of "futureproofing" drives me crazy. It may be nice if people saw mobos they liked and baught them to keep them in a closet for a couple years, but I believe most people buy their systems within a week or so. I can understand buying a 3000+ and maybe sticking in a $30 fx57 in three or four years, but you can do that with any board on the market pretty much. Adopting future standards before they can provide reasonable performance improvements will alter your financial experience far more than your computing experience. Then again I would have to blame the consumers more so than the companies as their tactics are obviously making them money. I would love to know how many SLI boards out there have one video card attached to them as well as how many unsued SATA channels there are out there. I just think it is dissapointing that in today's market its hard to find an extra feature you want without paying for five more that you don't.
_The object of the testing was to not only find the max 1:1 clockability of the board (310Mhz), but to find it's ability to reach the HTT limits of 2 different CPU's in our possession which certainly necessitates using a divider.
-This board booted into Windows at 510Mhz HTT. Having come from an extensive overclocking background, it is recognized that the fact a board will not boot into Windows at a specified speed does not necessarily mean the board/CPU will not function at an increased HTT speed. In this case, to determine the maximum HTT the Expert and 3500+ Winchester could attain the use of the Popular overclocking program ClockGen, and it only resulted in an additional 2Mhz HTT.
There are many readers that indeed take great interest in the overall top HTT overclockability of a board.
- The change in CPU's was merely to find the boards ability to overclock the HTT to it's maximum. This gem of a 3500+ was known to have reached an HTT of 500Mhz previously. The 4000+ has a maximum HTT of 400Mhz. Of course for all tests other than the Extreme Overclocking section, the 4000+ was utilized fror straight up comparisons.
-I have updated the gaming performance graphs which includes comparison to the A8N32-SLI
I'll update the other graphs as time allows.
I thank you wholeheartedly for your comments, Randi :)
Watercooling on the standard overclocking results??? I can see it on the extreme overclocking results but how many people watercool their systems in general and why not show air cooling. All other AT reviews are based on air-cooling, why the change now.
The gaming and general performance charts just do not make sense without comparing apples to apples. The 3Dmark chart is most confusing. Why not show some real game benchmarks, who freaking plays aquamark and pcmark?
Workstation performance??? WTF??? It was nice to see the AMD system romp on the Intel systems but if this board is for the extreme overclocker why do they care about Maya results?
Why compare 7800GTX results to 6800u results with different drivers? WTF is going on with this kind of testing.
Even though the audio/disk/ethernet performance should be the same, why not test it and see if something is different just in case. Considering the additional audio information in the last AMd and Intel board reviews, why not have it here with real games. Who knows, the cpu utilization could be a lot lower than other ALC850s (like Abit)or higher like Gigabyte.
quote: For this evaluation, we utilized a large selection of DDR memory.
Any reason not to mention how many and what the results were? Did Corsair or others have input into the testing like OCZ did? What kind of memory should I buy for this board?
It was good to see the overclocking information and what can be done with a board but why not show some real results with it. Can I play BF2 or COD2 at 320HTT? Please follow up with some game benchmarks at both standard and overclocked settings.
Randi's earlier review followed standard AnandTech motherboard testing, but we asked him on this review to go wherever he wanted in overclocking on this Expert board. We had already tested the original DFI nF4 SLI-DR, and the Expert was primarily an upgreade to provide even better overclocking performance and stability at high overclocked speeeds.
This seemed the perfect board to approach from an extreme overclocker's perspective - and that's exactly what Randi did. We decided a lot of water-cooling users would be interested in this board, so Randi used watercooling in some of his tests. He also included base performance numbers so you can compare to other AT reviews.
We have limited our standard reviews to air cooled overclocking, since that is very widely used these days (often by users who don't have a clue what they're doing). However, Gary, myself, and in particular Randi are all long-term overclockers, and sometimes we like to share a little more about where you can go with some of the equipment we test.
Consider this an extreme overclocker's review of an updated board clearly aimed at the extreme overclocker.
"Any reason not to mention how many and what the results were? Did Corsair or others have input into the testing like OCZ did? What kind of memory should I buy for this board?"
All the memory tested in the review was listed or mentioned(BH-5 based Kingston KHX3500 + OCZ Platiminum LE PC3500). The Kingston and OCZ BH-5 did not perform well on this board though the Corsair 3200LL BH-5 did and was included.The main reason for their poor performance could be attributed their age(2+ years old) and the fact I've pushed 3.6v+ through them in overclocking endeavors in the past.
OCZ did NOT have input into my memory testing, but their Guru Tony Leach provided some insight into the boards revisions as conveyed to him by DFI.
All results were listed as per the tables provided to illustrate top memory clocks for each of the Memory module sets.
As explained in my previous reply, I arrived here to review bringing with me an extensive overclocking background.The addition of watercooling brings a new dimension to the testing and evaluation performed in my reviews for AnandTech. There are many more people watercooling these days and its safe to to say that it is becoming more mainstream than even a couple of years ago.Of course when running a CPU at stock speeds in a review there is no benefit(outside of lower CPU temperature), the benefit only coming when attaining a top CPU overclock.
I also have several Phase Change coolers that someday may be used for the Top Overclock
section which could include Top Overclock on air, water and or Phase Change. Just a thought for now:)
As mentioned before, I did update the Gaming Performance charts and will update the others as time allows.
As far as Workstation performance, though this board is aimed at the Overclocking crowd, with it's full feature set it indeed can fulfill a Workstation role and does so admirably IMHO.
I believe the Corsair CMX1024-4000PT modules use the Samsung UCCC modules, not the Samsung TCCD modules. This would explain why they do not work at 2-2-2-5 latency.
Also, TCCD ICs do not come in 64x8 density, so it's impossible to make a 1GB TCCD module without somehow cramming 32 chips on there.
"With the SIL 3132 available, it would have made sense to include that SATAII controller, which would have given this board 8 SATAII channels and thus keeping this board future-proof and flexible."
How is a board with eight SATA connectors more future-proof than one with only 4 SATA connectors? Would anyone have more than 4 disk devices (hard drive, optical drive)?
Anyway, there are other four SATA ports from the chipset, so if you think they are lower performance, use them for SATA optical devices (when they will appear), and use the four connectors from the Sil controller to drive four hard drives.
The Expert uses the nForce4 chipset, so it supports SATA2 and NCQ. The performance of the nForce4 chipset is already well-documetned. THere are links on the first page to tests of all these nForce4 controllers and features.
Besides being too expensive in this particular case.. everyone should looking into an overclocking "complicated" board simply because they are built to last overclocked! Which means they should last longer stock than cheezy boards even if you're not interested in tweaking at all. Notice he mentions highend componets on board like Jap caps..better cooling etc.
I agree that good components and a stability are where it all begins.
Still, for a price like this, I'd miss the features AT mentions and Firewire 800a
(on a PCIe lane, please). Plus I don't care for SLI. A simple 16x 4x 1x PCIe would do.
To me this an overture to what DFI can do. This one is a bit too much in the OC niche for me.
What would you consider a cheezy board? Just curious.
Nice...Any word about a ultra expert? $200 way to much for a mobo if you don't need SLi. I can identify with heat on old ultra..mosfet HS get waaay too hot.. Another feature I like of this board is it looks like you might be able to replace fan with a passive Zalman thingi... old boards set chipset right under PCIe #1 leaving you no alternative but running a fan.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
40 Comments
Back to Article
Hardass1 - Sunday, February 19, 2006 - link
Another well done review Sir.Hardass.
lopri - Saturday, November 26, 2005 - link
If you had to pick one, which one would you pick? A8N32-SLI or Expert? Please don't tell me "Both are good", "They're different animals" or anything in that sense. Most of us have to pick ONE and that exact question is what we want to know. I'd venture to say what matters are following two:1. Overclockability
2. Stability
Fetures, layouts, etc... yeah.. they are all good and nice, but what really matters are whether the overclock is stable. And that's what brought DFI here today. Could you comment on it? If you had a choice to pick JUST ONE, which one would you pick?
Great review anyway. I'm actually happy that AT is becoming more enthusiast-friendly and looking forward to the next review. (Possibly Opteron Overclocking review?)
Thanks.
lop
RSica - Sunday, November 27, 2005 - link
Hi Lop:)I'll be honest in telling you I have not had my hands on the Asus board to know it's full overclocking abilities, so I really can't give you an opinion on which board to choose. Wesley could give you full insight on that one.
I go way back in the overclocking business, having hooked up with my best buds OPPAINTER and DDTUNG back in the day. At that time we were modding and overclocking each generation of Abit AMD based boards starting with the KG7 and culminating with the NF7-S, which we were pre-production testing prior to it's retail release. I've also had a play with Epox, Gigabyte, and the famous Shuttle A64 board.When I recieved the original SLI-D back in January it was an overclockers dream come true.All the voltage and overclocking/memory options and it overclocked way easy compared to the norm.(Thank you Oskar Wu:))
The Expert is more of the same but much better. I am a bit biased on a personal level about the board, and if you had just asked about it, I'd have given a thumbs up.
Thanks for your comments !
Randi
Scrith - Wednesday, November 30, 2005 - link
Does this use the same chipset as the A8N32-SLI? If not, why not, and where are the competitors for that board?Heckler 5th - Sunday, December 4, 2005 - link
how come the box the reviewer received already has that "anandtech gold medal" sticker on there? hmmm, kinda fishy... LOLcbkia - Saturday, November 26, 2005 - link
Under the extreme oc page, the pic showing the HTT @ 400MHz and RAM @ 300MHZ 2.5-3-3-8 but the sisoft is only 6674 MB/s? 300MHz should be displaying something near 8GB/sRSica - Sunday, November 27, 2005 - link
That would be true if not running at 2400Mhz.All HTT and Memory overclocking tests were done at a reduced multiplier. Each multiplier also has an effect on system bandwidth in the way the A64 responds to them.If I had chosen to run 7 and a total CPU Mhz of 2800, you would have most likely seen the 8000mb/s figure.In contrast, the stock 4000+ at 12x200Mhz with tight timings will average 5600-5700mb/s.
Another consideration is that with the bios used which made it more ram overclocking friendly, that there was possibly a relaxation of some of the bios register settings, which can also reduce bandwidth a bit in the name of pushing the memory higher in Mhz.
Thanks for your comments !
Randi
RobFDB - Saturday, November 26, 2005 - link
What i'm really interested in is if this board suffers from the same problem as my Ultra-D with PC4000 VX memory (the infamous cold boot issue)?I don't really want to pick this board up and have it suffer from the same problem. Not that i've tried my VX with the latest beta bios, but that's besides the point.
RobFDB - Tuesday, November 29, 2005 - link
Can anyone confirm or deny that the cold boot issue exists with this board?yacoub - Saturday, November 26, 2005 - link
If it had passive cooling like the A8N32-SLI I'd be more interested.karioskasra - Saturday, November 26, 2005 - link
Agreed. The nb fan on the dfi sounds like a cat getting mauled. It's enough to make you go water just for the quiet.karioskasra - Saturday, November 26, 2005 - link
I'm calling for dual cores to be included in the standard test setup.Bozo Galora - Saturday, November 26, 2005 - link
newegg has them in stock - 2 billsdecptt - Friday, November 25, 2005 - link
I wanna ask the author.Have you tested 4x1GB yet?
Can it see 4GB or can't it?
What timing do you use?
RSica - Saturday, November 26, 2005 - link
I did not test with 4x 1GB sticks because I only have 2 in my possession for reviewing. I did in fact as mentioned run 4x512MB sticks at 2T with no problem whatsover for testing.Future reviews will very well see x2 A64's included but I'm not sure what the timetable will be for their inclusion.
Again, thank you for your comments !
Randi
cryptonomicon - Saturday, November 26, 2005 - link
i think the article said that it couldn't pass 1T for 4x1gb, but it can run it (at 2t).SignalPST - Friday, November 25, 2005 - link
Thanks Randi for the very detailed review of the Expert.I was wondering, with the new layout of the RAM slots and CPU, is it possible to populate all 4 RAM slots and still be able to fit the thermalright XP-120 without problems?
tjr508 - Friday, November 25, 2005 - link
Bravo to DFI for using decent power supplies and such, but why is the SIL 3114 such a dissapointment? Does any desktop user need SATA2 at this time? Does anyone need DDR2 as opposed to high quality DDR (not this board but in a couple months). This whole idea of "futureproofing" drives me crazy. It may be nice if people saw mobos they liked and baught them to keep them in a closet for a couple years, but I believe most people buy their systems within a week or so. I can understand buying a 3000+ and maybe sticking in a $30 fx57 in three or four years, but you can do that with any board on the market pretty much. Adopting future standards before they can provide reasonable performance improvements will alter your financial experience far more than your computing experience. Then again I would have to blame the consumers more so than the companies as their tactics are obviously making them money. I would love to know how many SLI boards out there have one video card attached to them as well as how many unsued SATA channels there are out there. I just think it is dissapointing that in today's market its hard to find an extra feature you want without paying for five more that you don't.Tanclearas - Friday, November 25, 2005 - link
I had to double-check that I was indeed still reading a review at Anandtech.Since when is the max FSB of a motherboard determined using any memory ratio other than 1:1?
How many other boards reviewed at AT were booted into Windows at a certain FSB before being cranked up?
Why the change in CPU from previous tests (using a 3500 rather than 4000)?
What is with the graph for Futuremark that compares the different component scores of the same board? What is the point of graphing those numbers?
Where is the comparison to the A8N32-SLI? I'm pretty sure that's the benchmark board right now for enthusiasts.
I didn't even finish reading the review.
RSica - Friday, November 25, 2005 - link
Good Morning :)_The object of the testing was to not only find the max 1:1 clockability of the board (310Mhz), but to find it's ability to reach the HTT limits of 2 different CPU's in our possession which certainly necessitates using a divider.
-This board booted into Windows at 510Mhz HTT. Having come from an extensive overclocking background, it is recognized that the fact a board will not boot into Windows at a specified speed does not necessarily mean the board/CPU will not function at an increased HTT speed. In this case, to determine the maximum HTT the Expert and 3500+ Winchester could attain the use of the Popular overclocking program ClockGen, and it only resulted in an additional 2Mhz HTT.
There are many readers that indeed take great interest in the overall top HTT overclockability of a board.
- The change in CPU's was merely to find the boards ability to overclock the HTT to it's maximum. This gem of a 3500+ was known to have reached an HTT of 500Mhz previously. The 4000+ has a maximum HTT of 400Mhz. Of course for all tests other than the Extreme Overclocking section, the 4000+ was utilized fror straight up comparisons.
-I have updated the gaming performance graphs which includes comparison to the A8N32-SLI
I'll update the other graphs as time allows.
I thank you wholeheartedly for your comments, Randi :)
nvidia4ever1 - Friday, November 25, 2005 - link
Let's add the following-Watercooling on the standard overclocking results??? I can see it on the extreme overclocking results but how many people watercool their systems in general and why not show air cooling. All other AT reviews are based on air-cooling, why the change now.
The gaming and general performance charts just do not make sense without comparing apples to apples. The 3Dmark chart is most confusing. Why not show some real game benchmarks, who freaking plays aquamark and pcmark?
Workstation performance??? WTF??? It was nice to see the AMD system romp on the Intel systems but if this board is for the extreme overclocker why do they care about Maya results?
Why compare 7800GTX results to 6800u results with different drivers? WTF is going on with this kind of testing.
Even though the audio/disk/ethernet performance should be the same, why not test it and see if something is different just in case. Considering the additional audio information in the last AMd and Intel board reviews, why not have it here with real games. Who knows, the cpu utilization could be a lot lower than other ALC850s (like Abit)or higher like Gigabyte.
Any reason not to mention how many and what the results were? Did Corsair or others have input into the testing like OCZ did? What kind of memory should I buy for this board?
It was good to see the overclocking information and what can be done with a board but why not show some real results with it. Can I play BF2 or COD2 at 320HTT? Please follow up with some game benchmarks at both standard and overclocked settings.
Wesley Fink - Friday, November 25, 2005 - link
Randi's earlier review followed standard AnandTech motherboard testing, but we asked him on this review to go wherever he wanted in overclocking on this Expert board. We had already tested the original DFI nF4 SLI-DR, and the Expert was primarily an upgreade to provide even better overclocking performance and stability at high overclocked speeeds.This seemed the perfect board to approach from an extreme overclocker's perspective - and that's exactly what Randi did. We decided a lot of water-cooling users would be interested in this board, so Randi used watercooling in some of his tests. He also included base performance numbers so you can compare to other AT reviews.
We have limited our standard reviews to air cooled overclocking, since that is very widely used these days (often by users who don't have a clue what they're doing). However, Gary, myself, and in particular Randi are all long-term overclockers, and sometimes we like to share a little more about where you can go with some of the equipment we test.
Consider this an extreme overclocker's review of an updated board clearly aimed at the extreme overclocker.
RSica - Friday, November 25, 2005 - link
"Any reason not to mention how many and what the results were? Did Corsair or others have input into the testing like OCZ did? What kind of memory should I buy for this board?"All the memory tested in the review was listed or mentioned(BH-5 based Kingston KHX3500 + OCZ Platiminum LE PC3500). The Kingston and OCZ BH-5 did not perform well on this board though the Corsair 3200LL BH-5 did and was included.The main reason for their poor performance could be attributed their age(2+ years old) and the fact I've pushed 3.6v+ through them in overclocking endeavors in the past.
OCZ did NOT have input into my memory testing, but their Guru Tony Leach provided some insight into the boards revisions as conveyed to him by DFI.
All results were listed as per the tables provided to illustrate top memory clocks for each of the Memory module sets.
Again, thanks :)
Randi
RSica - Friday, November 25, 2005 - link
As explained in my previous reply, I arrived here to review bringing with me an extensive overclocking background.The addition of watercooling brings a new dimension to the testing and evaluation performed in my reviews for AnandTech. There are many more people watercooling these days and its safe to to say that it is becoming more mainstream than even a couple of years ago.Of course when running a CPU at stock speeds in a review there is no benefit(outside of lower CPU temperature), the benefit only coming when attaining a top CPU overclock.I also have several Phase Change coolers that someday may be used for the Top Overclock
section which could include Top Overclock on air, water and or Phase Change. Just a thought for now:)
As mentioned before, I did update the Gaming Performance charts and will update the others as time allows.
As far as Workstation performance, though this board is aimed at the Overclocking crowd, with it's full feature set it indeed can fulfill a Workstation role and does so admirably IMHO.
Again, thank you for your comments,
Randi
Griswold - Friday, November 25, 2005 - link
I love my Ultra-D but 2 things I cant stand:1. The name "Lanparty"
2. The ridiculous childish box
ceefka - Friday, November 25, 2005 - link
Yeah that kid doesn't seem to grow up with you ;-)Barbarossa - Friday, November 25, 2005 - link
I believe the Corsair CMX1024-4000PT modules use the Samsung UCCC modules, not the Samsung TCCD modules. This would explain why they do not work at 2-2-2-5 latency.Also, TCCD ICs do not come in 64x8 density, so it's impossible to make a 1GB TCCD module without somehow cramming 32 chips on there.
Live - Friday, November 25, 2005 - link
I beleive this is correct. If not this would be the worlds first 1gb stick with TCCD.Calin - Friday, November 25, 2005 - link
"With the SIL 3132 available, it would have made sense to include that SATAII controller, which would have given this board 8 SATAII channels and thus keeping this board future-proof and flexible."How is a board with eight SATA connectors more future-proof than one with only 4 SATA connectors? Would anyone have more than 4 disk devices (hard drive, optical drive)?
Anyway, there are other four SATA ports from the chipset, so if you think they are lower performance, use them for SATA optical devices (when they will appear), and use the four connectors from the Sil controller to drive four hard drives.
MarkHark - Friday, November 25, 2005 - link
More important than the number of available ports is features. What about NCQ support? Is the current controller fully SATA-II compatible?Wesley Fink - Friday, November 25, 2005 - link
The Expert uses the nForce4 chipset, so it supports SATA2 and NCQ. The performance of the nForce4 chipset is already well-documetned. THere are links on the first page to tests of all these nForce4 controllers and features.bob661 - Friday, November 25, 2005 - link
Nice board for the OCers. Too complicated for me.Zebo - Friday, November 25, 2005 - link
Besides being too expensive in this particular case.. everyone should looking into an overclocking "complicated" board simply because they are built to last overclocked! Which means they should last longer stock than cheezy boards even if you're not interested in tweaking at all. Notice he mentions highend componets on board like Jap caps..better cooling etc.ceefka - Friday, November 25, 2005 - link
I agree that good components and a stability are where it all begins.Still, for a price like this, I'd miss the features AT mentions and Firewire 800a
(on a PCIe lane, please). Plus I don't care for SLI. A simple 16x 4x 1x PCIe would do.
To me this an overture to what DFI can do. This one is a bit too much in the OC niche for me.
What would you consider a cheezy board? Just curious.
ceefka - Friday, November 25, 2005 - link
bummer, typo: 800a (?) forget the "a" please.Pete84 - Friday, November 25, 2005 - link
Wow, what an overclock!!bob661 - Friday, November 25, 2005 - link
Yeah. My jaw was dropped on that one. Never seen 500MHz on memory before. At least, not without extreme cooling.ViRGE - Friday, November 25, 2005 - link
The memory is not running at 500mhz, it was divided down to 250mhz. The 500mhz mark is purely a measure of the highest FSB that could be attained.NullSubroutine - Friday, November 25, 2005 - link
is this another penis test?Zebo - Friday, November 25, 2005 - link
Nice...Any word about a ultra expert? $200 way to much for a mobo if you don't need SLi. I can identify with heat on old ultra..mosfet HS get waaay too hot.. Another feature I like of this board is it looks like you might be able to replace fan with a passive Zalman thingi... old boards set chipset right under PCIe #1 leaving you no alternative but running a fan.