1366 x58 Xeon Enthusiast overclocks club

chessmyantidrug said:
Actually, there were i5 processors, but they were on the LGA 1156 socket. They were mainstream processors whereas i7-920 on up were enthusiast processors on a different socket. It's the same thing today, but the sockets are LGA 1150 (mainstream) and LGA 2011-3 (enthusiast).
Click to expand...
I remember it well.

I was going to get my wife an i5 750/P55 build but she insisted on an i7 920/X58 build.










And she was right.
 
I feel like enthusiast builds were more affordable back then. Then again, I got my i7-930 for $216 at Microcenter, which was about $124 less than they were on Newegg. I got my Sabertooth with an open box deal on Newegg for $93.
 
Just put my X5670 in today. Bought it for 12000yen (~$100USD).

After loading into Win7 x64 Pro, 4GB of my ram wasn't registering in Windows, but was registering in CPU-Z. I re-seated the CPU and loosened the heatsink a bit, and that fixed it. One of the pads must not have been touching a pin?

I'm only running at 4.0GHz right now. It is rock-solid stable at 1.248V core voltage after half a day on prime95 v28.5 so far. I can probably go lower on Vcore, or up the speed to 4.1 or 4.2GHz at that voltage.
qdzKpEi.jpg


I'll probably turn it up to 4.4GHz after the summer is over, but I need at least 1.35V at that to not code 0x101 BSOD. Summers here are very hot, the homes have no insulation, and A/C is expensive...
 
Last edited:
chessmyantidrug said:
I feel like enthusiast builds were more affordable back then. Then again, I got my i7-930 for $216 at Microcenter, which was about $124 less than they were on Newegg. I got my Sabertooth with an open box deal on Newegg for $93.
Click to expand...

I was lucky to grab an i7 920 for $150 off Craigslist back then though it was a C0 one. X58A-UD5 for $300 or so, XFX 5850 BE for $400+ (a 2nd one off Craigslist for $200 something 8 or so months later), 6 x 2GB of DDR3 for ~$420 ($200 for a 6GB kit off Craigslist that were defective but I got new replacements from G Skills) and a D14 for $50 (was a smoking deal then).

The budget for my wife's build went out the window since she had to have a Lian Li case too.
 
Last edited:
I spent a small fortune for my memory, too. Two Mushkin 998691 kits. I spent around $430 combined on them. At least they overclock really well, even with all six slots populated.
 
I might have 200 in all 12 gigs of ram im using....but mine wont over clock either....if i was positive it would make a big performance difference i might be tempted to upgrade it but from this point on out its all about keeping our investment low since ddr4 came out
 
primetime said:
I might have 200 in all 12 gigs of ram im using....but mine wont over clock either....if i was positive it would make a big performance difference i might be tempted to upgrade it but from this point on out its all about keeping our investment low since ddr4 came out
Click to expand...

Same here. I would really like to replace these sticks with a 3x4GB kit to put a little less stress on the IMC, but I know my next rig won't use DDR3. I honestly don't plan on upgrading for another two or three generations at least.
 
Killerxp100 said:
Most games run better with HT off because they depend on single thread performance. Everyone says how well GTA V is for multi-thread but I think its only really using up to 6 threads, maybe 8. I see roughly 15-20fps increase going from 4 cores to 6, but a 5 fps decrease using 12 threads.
Click to expand...

it is a little more complicated than that. I am time constrained right now so this Wiki will have to do, but there are much more in depth info on the negative impact of hyperthreading. There is an issue with latency as well, but i well threaded and predictable situations it isnt a draw back.

http://en.wikipedia.org/wiki/Hyper-threading#Drawbacks

Overall the performance history of hyper-threading was a mixed one in the beginning. As one commentary on high performance computing from November 2002 notes:

Hyper-Threading can improve the performance of some MPI applications, but not all. Depending on the cluster configuration and, most importantly, the nature of the application running on the cluster, performance gains can vary or even be negative. The next step is to use performance tools to understand what areas contribute to performance gains and what areas contribute to performance degradation.[12]

As a result, performance improvements are very application-dependent;[13] however, when running two programs that require full attention of the processor it can actually seem like one or both of the programs slows down slightly when Hyper-Threading Technology is turned on.[14] This is due to the replay system of the Pentium 4 tying up valuable execution resources, equalizing the processor resources between the two programs which adds a varying amount of execution time. The Pentium 4 "Prescott" and the Xeon "Nocona" processors received a replay queue, which reduces execution time needed for the replay system. This is enough to completely overcome that performance hit.[15]

According to a November 2009 analysis by Intel, performance impacts of hyper-threading result in increased overall latency in case the execution of threads does not result in significant overall throughput gains, which vary[13] by the application. In other words, overall processing latency is significantly increased due to hyper-threading, with the negative effects becoming smaller as there are more simultaneous threads that can effectively use the additional hardware resource utilization provided by hyper-threading.[16] A similar performance analysis is available for the effects of hyper-threading when used to handle tasks related to managing network traffic, such as for processing interrupt requests generated by network interface controllers (NICs).[17] Another paper claims no performance improvements when hyper-threading is used for interrupt handling.[18]
Click to expand...

There are better sources out there. Hyperthreading is not the same thing as having extra cores. It is not always a positive gain, true cors are always better. But in well threaded and predicable crunching, the results are impressive. Encoding and canned MT bechmarks, absolutely. Random action game environments, hyperthreading isnt so useful. Not unless you are using dual core that is completely overwhelmed, then hyperthreaading can and does have a positive effect. quads and six cores, it is usually a mixed bag.
 
I will be taking mine into windows 10 as I have the EVGA X58 3 way SLi board and hope to try some of that Nvidia/AMD combos out as I have a R9-280 and 770GTX to try..
 
Been away for quite some time and come back to see this. Seems intriguing. I'm currently running an i7 920 with a P6T Deluxe V2 (latest BIOS 1202) and 12 gigs (3x4) in one of my main machines. Best OC I could ever get out of it is 3.4 with a 200 BCLK and x17 multiplier. Would dropping around $100 on an x5670 net a notable increase in performance? Might be a good way to stave off the next upgrade for a bit.
 
TKESlinky said:
Been away for quite some time and come back to see this. Seems intriguing. I'm currently running an i7 920 with a P6T Deluxe V2 (latest BIOS 1202) and 12 gigs (3x4) in one of my main machines. Best OC I could ever get out of it is 3.4 with a 200 BCLK and x17 multiplier. Would dropping around $100 on an x5670 net a notable increase in performance? Might be a good way to stave off the next upgrade for a bit.
Click to expand...

Yes, at 4ghz+ you'll notice a pretty large increase in CPU heavy applications (close to 2x the performance), it does of course depend on the application. :)

An example in Cinebench:
4.4ghz x5670 = ~1005
3.4ghz i7 920 = ~540
 
Killerxp100 said:
I've been looking at this on ebay. Would you only have to replace the bios chip for it to work? Could be a cheap solution for someone wanting to get a x58 board.

http://www.ebay.com/itm/ASUS-Rampag...504?pt=LH_DefaultDomain_0&hash=item487d97a128

Bios Chip : http://www.ebay.com/itm/BIOS-CHIP-A...302?pt=LH_DefaultDomain_0&hash=item3caa4d542e
Click to expand...

Argh. Avoid the seller selling these boards (keystonememory)
He's been nothing but a pain to deal with so far. Never guessed it by the amount of feedback he has.
 
As an eBay Associate, HardForum may earn from qualifying purchases.
I have to stop looking on eBay for boards. I have another EX58-UD3R v1.6 coming but this past Saturday I was staring at my phone watching an MSI X58M at $83 plus $15.xx for shipping counting down the seconds. I was so tempted to bid on it but I don't need another freaking board.
 
Hazard said:
I have to stop looking on eBay for boards. I have another EX58-UD3R v1.6 coming but this past Saturday I was staring at my phone watching an MSI X58M at $83 plus $15.xx for shipping counting down the seconds. I was so tempted to bid on it but I don't need another freaking board.
Click to expand...

I know the feeling. Been bidding on too many boards lately..:)

I may just end up getting a dual socket board for my paired ES x5670's as they run too hot overclocked anyway.
 
tbob22 said:
Yes, at 4ghz+ you'll notice a pretty large increase in CPU heavy applications (close to 2x the performance), it does of course depend on the application. :)

An example in Cinebench:
4.4ghz x5670 = ~1005
3.4ghz i7 920 = ~540
Click to expand...

I second that. Good thing is that 4GHz is pretty much guaranteed with any X5670 chip provided your cooling can handle it. I score 970 points in Cinebench R15 at 4.2GHz.
 
Last edited:
Why eBay when you have in house sale for Westmere , Asus Gene III and 12 GB of ram for low ass price...

Just search in For Sale / Trade: FS: x58 CPU/MOBO/RAM combo
 
Last edited:
tbob22 said:
Yes, at 4ghz+ you'll notice a pretty large increase in CPU heavy applications (close to 2x the performance), it does of course depend on the application. :)

An example in Cinebench:
4.4ghz x5670 = ~1005
3.4ghz i7 920 = ~540
Click to expand...

Why compare a quad core and hexacore at different frequencies? This doesn't really tell us anything as it's not an apples-to-apples comparison.
 
chessmyantidrug said:
Why compare a quad core and hexacore at different frequencies? This doesn't really tell us anything as it's not an apples-to-apples comparison.
Click to expand...

I believe he used those examples because the person asking the question was currently maxed out at the same speed....So he wanted to give an example of what to expect is all;)
 
primetime said:
I believe he used those examples because the person asking the question was currently maxed out at the same speed....So he wanted to give an example of what to expect is all;)
Click to expand...

Then a better demonstration would have been benching the Xeon at 3.4GHz. At least then you see a direct comparison.
 
chessmyantidrug said:
Then a better demonstration would have been benching the Xeon at 3.4GHz. At least then you see a direct comparison.
Click to expand...

This quote might help you better understand why it was done that way;)

TKESlinky said:
Been away for quite some time and come back to see this. Seems intriguing. I'm currently running an i7 920 with a P6T Deluxe V2 (latest BIOS 1202) and 12 gigs (3x4) in one of my main machines. Best OC I could ever get out of it is 3.4 with a 200 BCLK and x17 multiplier. Would dropping around $100 on an x5670 net a notable increase in performance? Might be a good way to stave off the next upgrade for a bit.
Click to expand...
 
chessmyantidrug said:
Then a better demonstration would have been benching the Xeon at 3.4GHz. At least then you see a direct comparison.
Click to expand...

As mentioned, 3.4ghz was the max TKESlinky was able to get out of their 920.

Even if the 920 was at 4.4ghz the Cinebench would still be dramatically lower than the x5670, around 680 or so.

Of course Cinebench scores are a best case scenario, applications that aren't as CPU intensive will see less benefit.
 
Well 4.4GHz isn't guaranteed so I still don't see the point in the comparison. I was able to get 3.8GHz with my i7-930 and maxed out at 4.32GHz with my X5670.
 
chessmyantidrug said:
Well 4.4GHz isn't guaranteed so I still don't see the point in the comparison. I was able to get 3.8GHz with my i7-930 and maxed out at 4.32GHz with my X5670.
Click to expand...

I never said it was guaranteed, but it is my max stable OC while keeping voltages reasonable. Even the worst x5670 should be able to do 4ghz at 1.325v.

The point is to simply state that it will be a significant upgrade from that below average max OC on that 920.

The average for the x5670 seems to be around 4.2ghz at 1.325v. The average for a 920 is probably in the 3.6ghz range for that voltage, although both my w3520 and 920 did about 4.1ghz at that voltage.

If you want to compare "average" OC's of the two cpus it would be more like this:
x5670 4.2ghz=~970
i7 920 3.6ghz=~550

If you want to compare them at the same clockspeed:
x5670 3.6ghz=~850
i7 920 3.6ghz=~550
X5670 4.4ghz=~1005
i7 920 4.4ghz=~680

Also, the ram speed and uncore make a pretty big difference in Cinebench, so that can affect the score up to 20-30 points.

Whatever kind of comparison I make I'm sure you'll have some kind of problem with it.
 
pretty sure my board doesn't fully support the Xeon, tired of messing with it so X5670 for sale over in FS/FT
 
dezertrat said:
pretty sure my board doesn't fully support the Xeon, tired of messing with it so X5670 for sale over in FS/FT
Click to expand...

That shitty board is 6 phase not to mention it doesnt even have a heatsink on a MOSFETs of coarse your OC is not going to be stable... BIOS probably is old and shit too. Instead of selling that beautiful chip, buy a better board and that MSI crap.........introduce it to the hammer! :D
 
Vlad_13 said:
That shitty board is 6 phase not to mention it doesnt even have a heatsink on a MOSFETs of coarse your OC is not going to be stable... BIOS probably is old and shit too. Instead of selling that beautiful chip, buy a better board and that MSI crap.........introduce it to the hammer! :D
Click to expand...

It does fine with the 130W 920 and has for 5 years where the 5670 is only 95W Thanks for your opinion....
 
dezertrat said:
runs fine without OC, theory is my board doesn't fully support 32nm chips, it locks uncore ratio and its wrong (2x vs 1.5)
Click to expand...

mine defaults to 2x as well.....but its happy there anyway;) but i can easily change it if need be
 
dezertrat said:
It does fine with the 130W 920 and has for 5 years where the 5670 is only 95W Thanks for your opinion....
Click to expand...


Under heavy OC X56XX pull a lot more than 95W but anyway, have fun for another 5 years, while we enjoy 4+GHz with 12 threads! Your welcome!
 
I have a p6t deluxe v2 on the way with an x5670. Will I need to do a bios update before it will be functional? Also can I run ecc memory? It's cheaper and i want to run 3 x 8gb sticks
 
msbbc833 said:
I have a p6t deluxe v2 on the way with an x5670. Will I need to do a bios update before it will be functional? Also can I run ecc memory? It's cheaper and i want to run 3 x 8gb sticks
Click to expand...

Should update to newest bios and pretty sure x58 cannot run ECC
 
msbbc833 said:
I have a p6t deluxe v2 on the way with an x5670. Will I need to do a bios update before it will be functional? Also can I run ecc memory? It's cheaper and i want to run 3 x 8gb sticks
Click to expand...

Chances are the bios is up to date, but you never know. Might as well grab a $10 w3520 for testing.

I don't think ECC will work properly on that board (if at all). The P6T WS Pro, P6T7 WS SuperComputer, and P6T6 WS Revolution do support unbuffered ECC though.

Edit: Scratch that. It does support unbuffered 1333mhz ECC as well, support may be hit or miss though. http://www.asus.com/Motherboards/P6T_Deluxe_V2/HelpDesk_QVL/
You could grab one stick and try it out, if it doesn't work then try another. :)
 
msbbc833 said:
I have a p6t deluxe v2 on the way with an x5670. Will I need to do a bios update before it will be functional? Also can I run ecc memory? It's cheaper and i want to run 3 x 8gb sticks
Click to expand...
Because i really have so little experience using ecc ram can you post a link to ebay or wherever people are selling this ram (3 x 8gb sticks) your thinking of using....Your not the first person to ask about this and if it ends up working for you other people might want to know. Also i was just curious how the price compared versus normal buffered ram
 
You must log in or register to reply here.
Back
Top