Is an upgrade needed for BD-Rebuilder?

[bobrap]

I'm currently running an Intel Core2 Duo E6750 with 2GB ram on a Abit IP35Pro. Just wondering if a processor upgrade (quad core) would be worth it. Would the speed difference be worth the money? What's the average encoding time? The few movie only that I did took about 12 hrs+-. Thanks.

[rack04]

I'm currently running an Intel Core2 Duo E6750 with 2GB ram on a Abit IP35Pro. Just wondering if a processor upgrade (quad core) would be worth it. Would the speed difference be worth the money? What's the average encoding time? The few movie only that I did took about 12 hrs+-. Thanks.

Without knowing your source and x264 input, comparing times is very difficult. For instance, with my Q6600 and 2GB ram on a Asus P5K Deluxe It takes approximately 16 hours to encode a Blu-ray disc. Though I try to perform "high quality" encodes.

"%x264_PATH%" --pass 2 --bitrate %DESIRED_BITRATE% --stats "%WORKING_DIRECTORY%\%OUTPUT_FILENAME%.stats"
--level 4.1 --keyint 24 --min-keyint 1 --ref 3 --mixed-refs --no-fast-pskip --bframes 3 --b-adapt 2 --b-pyramid --weightb
--direct auto --deblock -1:-1 --subme 7 --trellis 2 --partitions all --8x8dct --vbv-bufsize 30000 --vbv-maxrate 24000 --qcomp 0.5
--me umh --merange 24 --threads auto --thread-input --sar 1:1 --progress --no-psnr --no-ssim
--output "%WORKING_DIRECTORY%\%OUTPUT_FILENAME%-output.h264" "%INPUT_VIDEO%" --mvrange 511 --aud --nal-hrd

[Dark Shikari]

I'm currently running an Intel Core2 Duo E6750 with 2GB ram on a Abit IP35Pro. Just wondering if a processor upgrade (quad core) would be worth it. Would the speed difference be worth the money? What's the average encoding time? The few movie only that I did took about 12 hrs+-. Thanks.A quadcore with the same processor speed as a dual core will run x264, as one would expect, twice as fast.

[rendez2k]

My quadcore (@3.05ghz) completes a 2hr movie in around 6hrs

[Chefkoch_ico]

My Intel Quad (Q6600 at 3GHz) does the main movie 1st pass at 21fps and the second pass at 7fps. Thats with High Quality setting in BD-RB and recode to a BD25.

For a 112 minute movie in sum about 8 Hrs.

Fps with a Core i7 would interest me.

Bye

[Wombler]

My Intel Quad (Q6600 at 3GHz) does the main movie 1st pass at 21fps and the second pass at 7fps. Thats with High Quality setting in BD-RB and recode to a BD25.

For a 112 minute movie in sum about 8 Hrs.

Fps with a Core i7 would interest me.

Bye

This (http://www.tomshardware.co.uk/Intel-Core-i7-Nehalem,review-31404-32.html) isn't quite what you're looking for but it will give you an idea of relative speeds.


Wombler

[MikeyBK]

I'm currently running an Intel Core2 Duo E6750 with 2GB ram on a Abit IP35Pro. Just wondering if a processor upgrade (quad core) would be worth it. Would the speed difference be worth the money? What's the average encoding time? The few movie only that I did took about 12 hrs+-. Thanks.


I have the Abit IP35Pro as well, with an E6600 Core2Duo OC'ing @3.2 GHz, and I have done 25 GB movie onlys in around 8-9 hours. I'm sure your E6750 can OC A Bit as well...:p

Another member from the LGSI forums has a Core2Quad Q9650 and he gets similar size BluRays done in around the 7 hour range... so not twice as fast, but can shave off an hour or two.....

BTW, that A Bit mobo has a superb windows-based overclocking utility called UGuru... perhaps you already have seen/tried it??

[bobrap]

Thanks everyone for your responses/help. Definitely gives me something to think about. Could be just an excuse to put together another machine:p

@MikeyBK

I never played with UGuru. Seems like a good time. Could you PM me your recommended settings. Thanks.

[prettyboy85712]

My quadcore (@3.05ghz) completes a 2hr movie in around 6hrs

My i7 920(@2.67ghz) completes a 2hr 14min movie in about 4hr 40min.

[MikeyBK]

Thanks everyone for your responses/help. Definitely gives me something to think about. Could be just an excuse to put together another machine:p

@MikeyBK

I never played with UGuru. Seems like a good time. Could you PM me your recommended settings. Thanks.

Will do. I think that you could get that baby up to 3.2GHz, no problem.;)

[Chefkoch_ico]

My Intel Quad (Q6600 at 3GHz) does the main movie 1st pass at 21fps and the second pass at 7fps. Thats with High Quality setting in BD-RB and recode to a BD25.

For a 112 minute movie in sum about 8 Hrs.

Fps with a Core i7 would interest me.

Btw. also Trellis was enabled.

[writersblock29]

@Bobrap

My AMD Phenom 9950 tends to average 20-30fps first pass, 7-8fps second pass, on 1080p AVC encoding. DVD standard res (720 X 480), I can see speeds of 30-35fps running a one-pass CRF. That's not overclocking, and certainly not a boast: It reminds me of the old days of DVD backups, when it wouldn't be out of the question to kill 12 hours on a 2 hour movie. Your milage will vary according to what the resolution is, whether any resizing is being done, the settings of your encoder, and whether or not you're leaving the system alone while encoding. Not to mention background programs like automatic defragmenters, antivirus/anti spyware programs... the list of variables is nearly endless.

I'll usually set up a project just before going to bed on a work night. That way, my computer has all that night and most of the next day left alone to do its thing... and the impact on my own computer time is non-existant. It only really gets frustrating if an error is hit: You can't help but to dwell on how much time the system just sat there, waiting for you to tell it what to do, while you were miles away and unaware. Thankfully, I don't have very many of those moments, and the backups get done. Of course, I don't buy very many movies all at one time, so the load isn't enormous for me.

[Capsbackup]

I'm running a Intel Core 2 Duo T9300 @ 2.50GHZ, with 4 GB RAM, Vista 32-bit. I am trying my 1st BD-25 full movie backup of Batman Begins, High Quality - Slower Encode selected. Started at 14:41:04 on 12-29-08, and at 21:41:47 the final M2TS began encoding. Now on pass 2 at 69% current progress; 89.55% overall progress on 12-31-08 at 07:37:01.:(
Looks like 48+ hours on this machine...

[jdobbs]

Ugghh... sorry but 1080p just takes a long time and requires a lot of processor time. I can't imagine 48 hours, though. My 2.5GHz AMD dual core takes about 12-18 hours depending on the disc.

[laserfan]

My 3.2GHz P4 has taken 48 hours. I imagine even a Core 2 Duo could take that long depending on encoding options and background processes.

@Capsbackup take a look at your background stuff--maybe there's a particular process slowing you down. For my part I even created a .bat file to shut-down every non-critical process quick-like.

But even doing all that I find BD backups to be something I do very rarely e.g. when I think the kids/guests might really want to beat-up on a movie, or e.g. when I want to re-process because I need to move subtitles into a 2.35 frame (Curse of the Golden Flower most recently).

[jdobbs]

I can't imagine a Core 2 duo running anywhere even near 48 hours -- even if you set it to the highest quality you can. Think about it... for a two hour movie to take that long it would have to be encoding at less than 1 frame per second. If my computer ran that slow, I'd use it as a boat anchor.

With the latest versions of BD-RB and X264 my Phenom 9500 (which is the slowest quad on the planet) takes about 8-10 hours to do a typical full disc backup. I've seen other people with quads report movie-only backups in the 4 hour range.

[TheOfficeMaven]

My Core 2 Duo took about 40 hours to encode a movie only run of The Dark Knight down to BD-25 (Dell XPS 720 machine). If I recall correctly, BD-RB stated that it was encoding at 0.70 fps on the first pass. A bit extream for sure, but well worth it when all is said and done. I only wish that the HD audio tracks could be retained for BD-25 encodes.

Thanks for all of your hard work on this one jdobbs. :thanks:

[jdobbs]

Wow -- something sounds wrong, that's terrible. My AMD Dual Core running at 2.4GHz does about 10fps on the first pass of a two pass encode (on a 1080p source that I just tested)... and it does about 7fps on a single CRF pass on the same source. I don't do many encodes on that machine -- I'll let it run and see how it does on some diverse sources -- but I guarantee it is going to be several times higher than what you're seeing.

[Capsbackup]

Still chugging along, 1.49fps, 88.30% current progress, 96.03% overall. Still might finish before this year ends.(ha ha) Never had a full movie encode take so long either. Could it be I selected BD-25? My four previous encodes were set to BD-5, full movie back-up and they took about 24-26 hours. Backround programs running are Daemon tools(not using it though), McAfee,HP Digital imaging monitor and a few Dell apps,(Dell Support, finger print reader, touch pad, quickset, and Google desktop). Maybe next encode I will try to disable these and see if speed improves.

[turbojet]

The BD25 output could have something to do with it, as lower quants take longer to encode. But that still sounds too slow for the cpu. Is x264 using 100% cpu?

[GaPony]

Still chugging along, 1.49fps, 88.30% current progress, 96.03% overall. Still might finish before this year ends.(ha ha) Never had a full movie encode take so long either. Could it be I selected BD-25? My four previous encodes were set to BD-5, full movie back-up and they took about 24-26 hours. Backround programs running are Daemon tools(not using it though), McAfee,HP Digital imaging monitor and a few Dell apps,(Dell Support, finger print reader, touch pad, quickset, and Google desktop). Maybe next encode I will try to disable these and see if speed improves.

The time it takes, PC hardware speaking, is more about the CPU than the RAM. You won't gain anything by disabling those background programs, as BDRebuilder is very easy on memory, but extremely intensive on the CPU. Its possible that your CPU is overheating and stepping down. You might try using the utility of your choice to monitor CPU temps while making your copy.

[Sophocles]

A quadcore with the same processor speed as a dual core will run x264, as one would expect, twice as fast.


Actually it won't. Going from a single core to a dual core does render about twice the speed, but going from a dual core to quad core will result in diminishing returns. Even still it is better to have a fast quad core when re-encoding BD.

[Dark Shikari]

Actually it won't.Yes, in fact, it will.

[Sophocles]

Yes, in fact, it will.


No in fact it won't! Try it. I don't know of a single multi-threaded software of any kind anywhere that will. In theory it should but there are other factors that will prevent it from doing so. Going from a single core to a dual core will see a doubling of speed, but going from a dual core to a quad core will see only about a 40% increase.

My systems specs

Q9650@4.05Ghz
4 GHz of RAM

Let's just say that it's fast.

I've been playing with X.264 since our discussion in the DiVx h.264 decoding beta. It does use 4 cores well but the benefits from a dual core to a quad core are not double.

[Dark Shikari]

going from a dual core to a quad core will see only about a 40% increase.So the dozens of tests here are all made up? (http://www.techarp.com/showarticle.aspx?artno=520)

They show a 94% performance improvement from dual to quadcore (both 45nm Core 2s at the same clock speed).

[laserfan]

While we're arguing (talking) about speeds, I wonder if anyone here can clarify a couple things for me, a lowly (for now) P4 Pentium owner:

1. Isn't there a whole angle on x264 encoding which needs to be addressed i.e. in order to Encode a video into x264 there needs to be Decoding going-on of the original video, whether VC-1, h264, MPEG2, or what-have-you. This Decoding process seems to me to be completely hidden from view wrt CPU resources i.e. one can't see DirectShow working. Yes?

2. As for x264 Encoding, what is correct for my P4 which is 1 core but HyperThreading, --threads auto, and --threads input? This latter seems to be about dedicating a thread to Avisynth but I suspect it makes no difference to my P4?

[Sophocles]

It scales extremely well up to about 5 threads and dives off after that, perhaps because of the overhead of splitting operations between the two separate CPU dies.

A lot of has to do with front side bus and memory, but it is also an issue with software design. There is just no way that a quad core is able to double a dual core of the same architecture and speed. I have 5 systems (getting ready to build a new one) that are all very fast by any standard (E8400@3.8 GHz, E6600@3.2 GHz, E6700@3.5Ghz, Opteron 175@2.5 GHz, and of course the Q9650.

I run video over and over again for no other purpose than to find the fastest means of Re-encoding, I'm strange that way. I think that you guys have done a great job with X.264, it is always my first choice, although I do play around with the other flavors as well. The issue is scaling and your test wasn't long enough to really test anything, and then there is the like processor architecture issue. Do a movie with an E8500 EO stepping dual core and then do it again with an EO stepping (Q9550 or Q9650) at the same clock speed and you will not see it go from 8 hours to 4 hours encoding.

[Dark Shikari]

A lot of has to do with front side bus and memory, but it is also an issue with software design. There is just no way that a quad core is able to double a dual core of the same architecture and speed. I have 5 systems (getting ready to build a new one) that are all very fast by any standard (E8400@3.8 GHz, E6600@3.2 GHz, E6700@3.5Ghz, Opteron 175@2.5 GHz, and of course the Q9650.

I run video over and over again for no other purpose than to find the fastest means of Re-encoding, I'm strange that way. I think that you guys have done a great job with X.264, it is always my first choice, although I do play around with the other flavors as well. The issue is scaling and your test wasn't long enough to really test anything, and then there is the like processor architecture issue. Do a movie with an E8500 EO stepping dual core and then do it again with an EO stepping (Q9550 or Q9650) at the same clock speed and you will not see it go from 8 hours to 4 hours encoding.I removed the test because it was not accurate (testing threads, rather than testing cores). See the above link I inserted instead.

That is a long enough test done with far more rigor than anything you have done yourself.

[Sophocles]

laser

As for x264 Encoding, what is correct for my P4 which is 1 core but HyperThreading, --threads auto, and --threads input? This latter seems to be about dedicating a thread to Avisynth but I suspect it makes no difference to my P4?

You will have to ask Dark for some of that. Hyper threading only works to your advantage if whatever application that you're using leaves part of of your processor unused. If an encoder uses 100% of your processor in a single thread then hyper threading can't work.

[Sophocles]

That is a long enough test done with far more rigor than anything you have done yourself.


I wouldn't bet on that. LOL You have no idea how much time and money I spend on building computers. I erect a new computer for my purposes every two to three months. I just sold two of them or I would have a testing lab of 7.

Also, I have a small forum of knowledgeable PC geeks testing on a regular basis.

Shark

You're an intelligent guy so try a little surfing to find out just how well quad cores scales up from a dual core and you will find all kinds of percentages that never get anywhere close to 100%

[Dark Shikari]

I wouldn't bet on that. LOL You have no idea how much time and money I spend on building computers. I erect a new computer for my purposes every two to three months. I just sold two of them or I would have a testing lab of 7.

Also, I have a small forum of knowledgeable PC geeks testing on a regular basis.As compared to a sample size of a couple hundred computers?You're an intelligent guy so try a little surfing to find out just how well quad cores scales up from a dual core and you will find all kinds of percentages that never get anywhere close to 100%I don't think you should be the one telling me to "do a little surfing."

[Sophocles]

As compared to a sample size of a couple hundred computers?

The sample is only as good as those providing it. Members of my forum and I are uniquely qualified to produce superior results.


I don't think you should be the one telling me to "do a little surfing."



Perhaps not but if you did then you would discover that what I've been telling you is true. There is no such thing as 100% scaling from a dual core to a quad core. Any claims to the contrary are a myth.

Don't you think that I would rejoice if 100% scaling to more cores was a reality?


Perhaps we should start a new thread to pursue this point because it seems to be a discussion worth having.

[Dark Shikari]

it seems to be a discussion worth having.No it isn't. I am not going to bother debating with an obvious troll who claims to be "uniquely qualified" so as to be able to contradict all other results in existence. Having demonstrated sufficiently for myself that you are attempting to troll me, this will be my last post on the topic.

[GaPony]

Not to get too wrapped in all the technical specifications, but the reality, insofar as BD-Rebuilder is concerned, is that a quad core PC of a given speed, is not going to do anything twice as fast as a dual core PC of the same speed. I have converted some movies on both my E6850 (3.0ghz) and my Q9650 (3.0ghz) with very similar PCs built around them. The anecdotal evidence is that the Q9650 is about 25-30% faster in these particular operations.

I don't have a dog in this fight, and the discussion is interesting, but the reality is what it is. I'm not a technical specs kind of guy, I'm just a results oriented kind of guy, and results always speak for themselves. ;)

[Sophocles]

No it isn't. I am not going to bother debating with an obvious troll who claims to be "uniquely qualified" so as to be able to contradict all other results in existence. Having demonstrated sufficiently for myself that you are attempting to troll me, this will be my last post on the topic.


Nice attempt to dodge the fact (yes fact) that your original claim of 100% scaling from dual core to quad core is a myth, untrue, can't happen. Calling me a troll doesn't make me a troll or you right, it makes you glib. Check around and you will see that I am not a troll. I generally excuse myself from debates with individuals who know more about a particular subject than I do, in your case X.264, but you stepped through my door when you started making erroneous claims about hardware that you are not that knowledgeable about. If you want to find out which of us is right then let's stop slicing baloney and start cutting some real meat. It's obvious to me that a real test of X.264's scaling needs to be tested. Again I really am a big fan of X.264, you guys have done a wonderful job with it and I hope that you can indeed get it to scale to 100% with each core doubling, but it's not there now.

[laserfan]

I wonder if anyone here can clarify a couple things for me, a lowly (for now) P4 Pentium owner:

1. Isn't there a whole angle on x264 encoding which needs to be addressed i.e. in order to Encode a video into x264 there needs to be Decoding going-on of the original video, whether VC-1, h264, MPEG2, or what-have-you. This Decoding process seems to me to be completely hidden from view wrt CPU resources i.e. one can't see DirectShow working. Yes?

2. As for x264 Encoding, what is correct for my P4 which is 1 core but HyperThreading, --threads auto, and --threads input? This latter seems to be about dedicating a thread to Avisynth but I suspect it makes no difference to my P4?

You will have to ask Dark for some of that. Hyper threading only works to your advantage if whatever application that you're using leaves part of of your processor unused. If an encoder uses 100% of your processor in a single thread then hyper threading can't work.
The application is x264 of course (no gui). Do I use --threads input or not? :confused:

[Sophocles]

The application is x264 of course (no gui). Do I use --threads input or not?


Probably not, X.264 makes use of most of your processor running a single thread so there is nothing left for the processor to run a second thread. Hyper threading only worked with the unused portion of you processor so if there is nothing unused there is no Hyper threading.

[bobrap]

Wow, guys. Didn't mean to start an argument/discussion.:devil: Just wanted to know if it was worth the money for me to upgrade to a quad from a duo. I mean if I only save 2 hrs+/- encoding time, what's the point? If upgrade is "warranted", any processor suggestions? Thanks and play nice,:)

[Sophocles]

In my world 2 hours is still quite a bit of time saved. The discussion also involved like architecture. For instance, if we are comparing an E6700 to a Q6700 at the same clock speed you will not encode at twice the speed, but it will encode about 30% to 40% faster. If one is looking at a processor with different architectures say going from an E6700 to a Q9550 then even at the same speed as the Q6700 it's going to be a little faster.

The Q9550 also has a slightly higher clock speed which will make it a bit faster than the E6700 still. Now the latest Q9550 EO stepping processors are excellent and easy to overclock to 3.6 GHz. At that speed it will be about twice as fast as a default speed E6700. I have a Q9650 overclocked to 4 GHz and I am getting about twice the speed of an E6700 which I also have. While were discussing architecture there is also the new i7 processors which are clock per clock about 30% to 40% faster for encoding than the Q9550, and that should shave some serious time off.

[GaPony]

Wow, guys. Didn't mean to start an argument/discussion.:devil: Just wanted to know if it was worth the money for me to upgrade to a quad from a duo. I mean if I only save 2 hrs+/- encoding time, what's the point? If upgrade is "warranted", any processor suggestions? Thanks and play nice,:)

You ask the age old question.... "Should I upgrade?" :)

The age old answer... taken from racing, "Speed costs, how fast does your wallet want to go?"

You have to determine for yourself whether the time savings, whether a little or alot, is worth the expense. You can probably do a simple CPU ugrade for not alot of money. On the other hand you may need other components to really make good use of a reasonably fast quad, like the Q9550.

Your choice is especially difficult with the release of the Intel i7 series of CPU's. That isn't a cheap upgrade, but in tests I've seen, its remarkably good at tasks such as those required by this software.

I guess my advice to you, with the state of technology where it is... Go big or stay home. All else being equal, If you currently have a slower dual core and upgrade to a fast quad, you will see some improvement, but how much it cuts your times is hard to tell. The i7 should bring much more dramatic results, but at a much greater cost.. I can tell you that with a Q9650, 8gb RAM, and Vista Ultimate 64-bit, no movie has taken over 8 hrs, and most fall into the 5.5hr - 6.5hr time frame. I don't OC when doing this kind of work, so a slightly OC'd Q9550 should get about the same results.

I hope this helps you decide whats best for you..

[Sophocles]

I can tell you that with a Q9650, 8gb RAM, and Vista Ultimate 64-bit, no movie has taken over 8 hrs, and most fall into the 5.5hr - 6.5hr

How much do you have it overclocked. I have mine water cooled at 4 GHz and my times are on average less than 5 hours for movies in the 2 hour range.


Laserfan

The reason that I purchased the Q9650 was because it was the only way that I was guaranteed to get an EO stepping processor at the time. A Q9550 with EO stepping will easily overclock to 3.8 GHz even with a good air cooler. A large portion of the Q9550s' out there are C1 stepping which isn't as good, but if you go to Tank Guys they guarantee EO stepping for a reasonable price. I would highly recommend this processor if you're not thinking of going to i7 processors.

http://www.tankguys.com/processors/intel-processors/core-2/intel-core-2-quad-q9550-eo-stepping.html

[ron spencer]

I have been doing these with a Pentium D (remeber those LOL) speed 3.4 GHz

I have a CPU here I could put in (best my ASus mobo can do). It is a Core2Duo E4500 2.20 GHz 2MB L2 Cache, 800 MHz FSB. This alot better?

[Sophocles]

I have a CPU here I could put in (best my ASus mobo can do). It is a Core2Duo E4500 2.20 GHz 2MB L2 Cache, 800 MHz FSB. This alot better?

Which Asus board do you have? But yes a core2 duo should outperform your D series processor for encoding. It would be really helpful if you could name your exact board.

[ron spencer]

my board is about 1.5 years old...I think that old

ASUS P5PE-VM Motherboard

[GaPony]

How much do you have it overclocked. I have mine water cooled at 4 GHz and my times are on average less than 5 hours for movies in the 2 hour range.


Laserfan

The reason that I purchased the Q9650 was because it was the only way that I was guaranteed to get an EO stepping processor at the time. A Q9550 with EO stepping will easily overclock to 3.8 GHz even with a good air cooler. A large portion of the Q9550s' out there are C1 stepping which isn't as good, but if you go to Tank Guys they guarantee EO stepping for a reasonable price. I would highly recommend this processor if you're not thinking of going to i7 processors.

http://www.tankguys.com/processors/intel-processors/core-2/intel-core-2-quad-q9550-eo-stepping.html

I'm not overclocking this Q9650 at all for this. I thought it would be better to do the beta testing at stock speeds. I can easily OC to 4.25ghz and I'm sure that would speed things up a tad.. :) I use a ASUS P5Q Deluxe mobo.

Unless you're jumping to the i7, the Q9650 is probably the most reliable, best bang for the buck. It didn't hurt that I got mine from a buddy for $200... still in the box. He caught i7 fever in the middle of a build. :)

[Sophocles]

I can easily OC to 4.25ghz and I'm sure that would speed things up a tad.. I use a ASUS P5Q Deluxe mobo.

I'm also using a P5Q Deluxe with a Q9650 water cooled processor, but I don't think that I can hit 4.25Ghz stable, and run accurate tests. It does however hit 4.05GHz stable and is by any measurable standard completely stable within specifications. The overclock shaves hours (more than two) off most single 2 hour re-encodes. Can't wait for an 8 core processor.:P

[GaPony]

I never said it was stable at 4.25ghz.. :D Its lasts maybe a minute before it folds under OCCT. Running everything else it gets hot, but not damaging hot at that speed.

Its good at 3.8ghz with every test I've run over a 24hr period, with just the Zalman HSF.

Maybe I'll step it up a notch or two, once I have a catalog of times built up, in order to compare results.

[bobrap]

Just thought I'd follow up and say that I went for the q9550 eo stepping. However.....I can't get it to run correctly. Stock cooler and I keep getting boot errors .8.7. Tons of resets and I can get it to boot but runs HOT! Boot into bios, temp reads 103c! :confused: Think I might just get a new MB. Looking at GIGABYTE GA-EP45-UD3R. Didn't know my name was Murphy :p

[GaPony]

You might check to see if there's a mobo BIOS update that might improve things... The stock cooler often isn't enough. It depends alot on the mobo and air flow of the case.

[bobrap]

I'm using the latest beta bios and the case is wide open. 3 fans internally when closed (Antec P180). Guess just a good excuse for a new MB. Any recommendations for CPU coolers?