Monday, December 12th 2011

Intel Core i7-3960X and i7-3930K CPUs Transitioning to C2 stepping in January

As previously reported, Intel's first wave of Sandy Bridge-E processors have VT-d (Intel Virtualization Technology for Directed I/O) disabled due to an errata in the C1 stepping. That issue couldn't be resolved in time for the launch but it's getting fixed with the C2 stepping which is set to start rolling out to customers on January 20th, 2012 (samples have already been delivered).

The CPUs moving to the C2 stepping are the hexa-core Core i7-3960X (3.3 GHz) and Core i7-3930K (3.2 GHz). Beside the fixed VT-d, the C2 chips will feature new S-spec and MM numbers so a BIOS update for current motherboards will likely be required.
Add your own comment

59 Comments on Intel Core i7-3960X and i7-3930K CPUs Transitioning to C2 stepping in January

#26
radrok
Also I think the best OC will come out with D0, sorry for double post, I wanted to edit :)
Feel free to delete this.
Posted on Reply
#27
repman244
Live OR Diewhat will they do with all the old C1 CPUS LOL
Rebrand them and sell them back as C2 stepping? :roll::roll:
Posted on Reply
#28
Aquinus
Resident Wat-man
Weather you realize it or not, Windows 7 will virtualize individual processes to protect private application memory. Virtualization helps everyone if you realize it or not. The people who will find this the most useful will most likely be administrators and developers who run virtual machines. If you're getting LGA2011 for gaming, you're just an idiot because this platform has a lot fo raw power for applications that actually are multithreaded, and thread-for-thread, this is almost just as fast as a 2600K.
Posted on Reply
#29
cadaveca
My name is Dave
AquinusIf you're getting LGA2011 for gaming, you're just an idiot because this platform has a lot fo raw power for applications that actually are multithreaded, and thread-for-thread, this is almost just as fast as a 2600K.
I beg to differ. In multi-GPU gaming, I get far higher frames on my 3960X than my 2600K could ever hope to give me.

SKT2011 is high-end computing, for sure, but gaming with multiple VGAs and monitors needs that grunt and bandwdith too.


I'm sorry, but calling people idiots isn't exactly within forum rules. Technically, you'd be missing a big part of why 2011 is good for gaming too, so you might want to reconsider your statement.


And yes, I can back my claim up with benchmarks. I dropped 2600K and the G1.Sniper2 for gaming, although i do miss how quiet that rig was.

On a lighter note, I'm eager to try a new stepping, know of several peopel keeping an eye out for them, yet here we are at the end of January, and none to be found. I wanna pick up a 3930K. C'mon, Intel, lets get stock on shelves.!
Posted on Reply
#30
edgedemon
For those in the UK, OCUK are selling the C2 stepping for the 3930K as I bought one...
Posted on Reply
#31
Aquinus
Resident Wat-man
Go to Guru3d.com's website and check out their CPU/GPU Scaling review and you will see how wrong you are in 90% of all. Most games won't use that many threads and higher clocks speeds and lower memory latencies are going to give you the best gains. At high resolutions your taxing your GPU much harder than the CPU, so unless your planning on running games at low resolutions with SLI or Crossfire(...and why?), it won't help as much.

I have a degree in Computer Science and I'm a Systems Administrator. I like to think that I know what I'm talking about since I work with hardware on a daily basis, but any person who knows what they're talking about will tell you that SB-E is overkill for games as they are today.

I apologize for calling the author an idiot, but seriously, do some research before you start calling people out.

www.guru3d.com/article/radeon-hd-7970-cpu-scaling-performance-review/

...and to quote the review:
If anything, this little article proofs once again that investing money in a faster graphics card will gain you better game performance compared to investing in a faster CPU. The performance difference in-between a 1000 USD Core i7 3960X compared to a 320 USD Core i7 2600K processor is extremely small, something you'd never notice unless measured. So we say, stick to a modern mainstream quad-core processor and the differences really aren't that big in the overall framerate, especially at 1920x1080/1200. Yes we know it's that weird penumbra, the higher you go in resolutions, the slower your processor may be. Remember, once you pass 1920x1080/1200 the GPU is almost always the bottleneck, not your processor.
Edit: Also unless you're running crossfire 7970s, this is a very realistic perspective on CPU scaling as it is just as fast as many current multi-gpu solutions. Also, benchmarks are designed to push everything in your system hard, that is their purpose, I'm talking about games and real world applications.
cadavecaI beg to differ. In multi-GPU gaming, I get far higher frames on my 3960X than my 2600K could ever hope to give me.

SKT2011 is high-end computing, for sure, but gaming with multiple VGAs and monitors needs that grunt and bandwdith too.


I'm sorry, but calling people idiots isn't exactly within forum rules. Technically, you'd be missing a big part of why 2011 is good for gaming too, so you might want to reconsider your statement.


And yes, I can back my claim up with benchmarks. I dropped 2600K and the G1.Sniper2 for gaming, although i do miss how quiet that rig was.

On a lighter note, I'm eager to try a new stepping, know of several peopel keeping an eye out for them, yet here we are at the end of January, and none to be found. I wanna pick up a 3930K. C'mon, Intel, lets get stock on shelves.!
Posted on Reply
#32
PaulieG
AquinusGo to Guru3d.com's website and check out their CPU/GPU Scaling review and you will see how wrong you are in 90% of all. Most games won't use that many threads and higher clocks speeds and lower memory latencies are going to give you the best gains. At high resolutions your taxing your GPU much harder than the CPU, so unless your planning on running games at low resolutions with SLI or Crossfire(...and why?), it won't help as much.

I have a degree in Computer Science and I'm a Systems Administrator. I like to think that I know what I'm talking about since I work with hardware on a daily basis, but any person who knows what they're talking about will tell you that SB-E is overkill for games as they are today.

I apologize for calling the author an idiot, but seriously, do some research before you start calling people out.

www.guru3d.com/article/radeon-hd-7970-cpu-scaling-performance-review/

...and to quote the review:


Edit: Also unless you're running crossfire 7970s, this is a very realistic perspective on CPU scaling as it is just as fast as many current multi-gpu solutions. Also, benchmarks are designed to push everything in your system hard, that is their purpose, I'm talking about games and real world applications.
Thing is, this is a tech enthusiast site. Many of us want the additional power, additional eye candy, and we will push the envelope to squeeze every ounce we can. So, the argument of a "real world" perspective will fall on deaf ears here. It's not really a matter of need on a forum like this. So while in a sense, what you say is true, it really doesn't apply here. If you think about it, if all we thought about was need, most of us would still be on 775 quad systems with no more that a 6870 or 460 gpu. :laugh:
Posted on Reply
#33
radrok
The 3960X is probably slightly better on performance Core Vs Core because of the higher L3.
I wouldn't be surprised if this difference will be more marked when Intel releases a full 20MB L3 Ivy-E, also I don't think that if you pay more money for a handful of FPS you are an "idiot" the term is enthusiast and there is nothing to be afraid of, why must people judge what others do with their money/free time? As long as you are having fun then who cares :)
Posted on Reply
#34
qubit
Overclocked quantum bit
cadavecaOn a lighter note, I'm eager to try a new stepping, know of several peopel keeping an eye out for them, yet here we are at the end of January, and none to be found. I wanna pick up a 3930K. C'mon, Intel, lets get stock on shelves.!
I tell ya, don't just wait for C2, but wait for the fully enabled 8 core. :rockout: That'll be the mark of a true enthusiast (with deep pockets, hehe). :D
PauliegThing is, this is a tech enthusiast site. Many of us want the additional power, additional eye candy, and we will push the envelope to squeeze every ounce we can. So, the argument of a "real world" perspective will fall on deaf ears here. It's not really a matter of need on a forum like this. So while in a sense, what you say is true, it really doesn't apply here. If you think about it, if all we thought about was need, most of us would still be on 775 quad systems with no more that a 6870 or 460 gpu. :laugh:
Yup, so true. That's why I recently upgraded my perfectly functional E8500 for a 2700K. Moar power! Moar threads! bla bla bla. Nerdgasmic stuff. :)

It's also why, when I have a GTX 580, I've just bought a GT 520 2GB (are you mad?! they say. Well, possibly). The reason being that I'm curious to see how my 580 with it's memory maxed out compares to the weedy 520 at the same settings with its memory not maxed out. Also, I've got the top and the bottom of the 5xx range, which I like having, but don't tell anyone. ;)

Yeah, I don't need any of this, real world be damned. :p
Posted on Reply
#35
radrok
qubitI tell ya, don't just wait for C2, but wait for the fully enabled 8 core. That'll be the mark of a true enthusiast (with deep pockets, hehe).
I hope it'll be out soon :toast:
Posted on Reply
#36
Aquinus
Resident Wat-man
If you read my flaming post (sorry again,) but I did say if you're buying LGA2011 *just* for gaming. In that respect it's over-kill. The point I was making was against this statement:
I beg to differ. In multi-GPU gaming, I get far higher frames on my 3960X than my 2600K could ever hope to give me.
That is just simply not true, and that is what I was getting at. Don't get me wrong, I'm planning on getting a LGA2011 system. I know that a 2500K or 2600K in all realism would do what I need, but I would still like a 3930k and 32gb of ram anyways.
Posted on Reply
#37
qubit
Overclocked quantum bit
AquinusIf you read my flaming post (sorry again,) but I did say if you're buying LGA2011 *just* for gaming. In that respect it's over-kill. The point I was making was against this statement:



That is just simply not true, and that is what I was getting at. Don't get me wrong, I'm planning on getting a LGA2011 system. I know that a 2500K or 2600K in all realism would do what I need, but I would still like a 3930k and 32gb of ram anyways.
TPU wouldn't employ cadaveca as our hardware reviewer if he didn't know what he was talking about, so if he says that performance is better, then you can take that as the truth. Note that he did qualify that the difference really becomes apparent with multiple graphics cards in the system. Presumably, this is to get over the driver overheads of running them.

If you've tested a similar setup and didn't see such gains, then you would have a much more productive conversation if instead of telling him that he doesn't know what he's talking about, you 'compared notes' ie discussed the finer points of your two systems, such as things like specific BIOS settings, driver settings, overclocks and all that good stuff. Doing that would be likely to close the gap and lead to agreement quite quickly.

TPU is a fantastic forum and you'll get a lot out of it if you put the effort to treat people right and stay within the forum rules. :)

Welcome to TPU. :toast:
Posted on Reply
#38
radrok
Performance difference is there, probably mostly due to the additional lanes, even though I don't like using Tom's it was the first that came up to my mind.
Posted on Reply
#39
PaulieG
radrokPerformance difference is there, probably mostly due to the additional lanes, even though I don't like using Tom's it was the first that came up to my mind.
media.bestofmicro.com/0/A/314938/original/battlefield%203%201680.pngmedia.bestofmicro.com/0/B/314939/original/battlefield%203%201920.pngmedia.bestofmicro.com/0/C/314940/original/battlefield%203%202560.png
I knew there was some minimal difference, but nowhere near worth the price difference b/t platforms.
Posted on Reply
#40
radrok
Yeah, Ideally you'd be better off with an i7-3820 to gain the PCIe lanes and 24/7 single stage phase change (this if the chip overclocks like a 2600k), but that's extreme :D
Posted on Reply
#41
Aquinus
Resident Wat-man
qubitTPU wouldn't employ cadaveca as our hardware reviewer if he didn't know what he was talking about, so if he says that performance is better, then you can take that as the truth. Note that he did qualify that the difference really becomes apparent with multiple graphics cards in the system. Presumably, this is to get over the driver overheads of running them.

If you've tested a similar setup and didn't see such gains, then you would have a much more productive conversation if instead of telling him that he doesn't know what he's talking about, you 'compared notes' ie discussed the finer points of your two systems, such as things like specific BIOS settings, driver settings, overclocks and all that good stuff. Doing that would be likely to close the gap and lead to agreement quite quickly.

TPU is a fantastic forum and you'll get a lot out of it if you put the effort to treat people right and stay within the forum rules. :)

Welcome to TPU. :toast:
Thank you. I actually don't usually work with consumer hardware, I mostly work with servers so I usually see Xeons, so I just read reviews on them. I'm just saying that I'm seeing reviews that disagree with that statement. Now I don't have the information for running multiple GPUs, but in the past I've noticed that the EE and the extra cores doesn't usually perform that much better over the first few chips below it as far as games are concerned. Games just simply aren't multithread to that extent. Clock for clock, SB and SB-E are architecturally the same, with a little extra latency on the L3 on the SB-E EE because of the size (3 ns? It's not a lot.) So if you have an i7 2600k and a 3960X both clocked at the same frequency, I would be surprised to see how many games perform better with the 3960X. That's also keeping other components the same, such as memory (clearly the SB-E would run quad-channel memory, which is a plus, but for games isn't really necessary. It's more bandwidth, but that doesn't always mean faster.) I think I recall reading that the 3960x is actually two SB dies on the same chip, but with two cores disabled and laser cut.

I don't really want to drag this out, but I'm just saying that it doesn't add up and that I can't imagine the gains are really that noticeable and I see nothing to support the contrary.
Posted on Reply
#42
Aquinus
Resident Wat-man
AquinusThank you. I actually don't usually work with consumer hardware, I mostly work with servers so I usually see Xeons, so I just read reviews on them. I'm just saying that I'm seeing reviews that disagree with that statement. Now I don't have the information for running multiple GPUs, but in the past I've noticed that the EE and the extra cores doesn't usually perform that much better over the first few chips below it as far as games are concerned. Games just simply aren't multithread to that extent. Clock for clock, SB and SB-E are architecturally the same, with a little extra latency on the L3 on the SB-E EE because of the size (3 ns? It's not a lot.) So if you have an i7 2600k and a 3960X both clocked at the same frequency, I would be surprised to see how many games perform better with the 3960X. That's also keeping other components the same, such as memory (clearly the SB-E would run quad-channel memory, which is a plus, but for games isn't really necessary. It's more bandwidth, but that doesn't always mean faster.) I think I recall reading that the 3960x is actually two SB dies on the same chip, but with two cores disabled and laser cut.

I don't really want to drag this out, but I'm just saying that it doesn't add up and that I can't imagine the gains are really that noticeable and I see nothing to support the contrary.
...and at frame-rates that high, I consider <2fps increase at high resolutions minimal.
Posted on Reply
#43
qubit
Overclocked quantum bit
AquinusThank you. I actually don't usually work with consumer hardware, I mostly work with servers so I usually see Xeons, so I just read reviews on them. I'm just saying that I'm seeing reviews that disagree with that statement. Now I don't have the information for running multiple GPUs, but in the past I've noticed that the EE and the extra cores doesn't usually perform that much better over the first few chips below it as far as games are concerned. Games just simply aren't multithread to that extent. Clock for clock, SB and SB-E are architecturally the same, with a little extra latency on the L3 on the SB-E EE because of the size (3 ns? It's not a lot.) So if you have an i7 2600k and a 3960X both clocked at the same frequency, I would be surprised to see how many games perform better with the 3960X. That's also keeping other components the same, such as memory (clearly the SB-E would run quad-channel memory, which is a plus, but for games isn't really necessary. It's more bandwidth, but that doesn't always mean faster.) I think I recall reading that the 3960x is actually two SB dies on the same chip, but with two cores disabled and laser cut.

I don't really want to drag this out, but I'm just saying that it doesn't add up and that I can't imagine the gains are really that noticeable and I see nothing to support the contrary.
Oh, absolutely. SB-E has the same threading performance and the same core as SB, but Intel have just bolted on more cores and increased memory bandwidth to handle them, plus those benches have indeed shown performance as nearly the same between SB & SB-E. All I'm saying is that if cadaveca happens to have a configuration which shows a clear difference between the two, it would make for an interesting and productive conversation to see exactly why he's seeing it.
Posted on Reply
#44
cadaveca
My name is Dave
AquinusGo to Guru3d.com's website and check out their CPU/GPU Scaling review and you will see how wrong you are in 90% of all. Most games won't use that many threads and higher clocks speeds and lower memory latencies are going to give you the best gains. At high resolutions your taxing your GPU much harder than the CPU, so unless your planning on running games at low resolutions with SLI or Crossfire(...and why?), it won't help as much.

I have a degree in Computer Science and I'm a Systems Administrator. I like to think that I know what I'm talking about since I work with hardware on a daily basis, but any person who knows what they're talking about will tell you that SB-E is overkill for games as they are today.

I apologize for calling the author an idiot, but seriously, do some research before you start calling people out.

www.guru3d.com/article/radeon-hd-7970-cpu-scaling-performance-review/

...and to quote the review:


Edit: Also unless you're running crossfire 7970s, this is a very realistic perspective on CPU scaling as it is just as fast as many current multi-gpu solutions. Also, benchmarks are designed to push everything in your system hard, that is their purpose, I'm talking about games and real world applications.
First, I'm not going to knock another review site. But what I will say is that perhaps I have a different perspective, and different usages than what some people employ. That doesn't mean either of us is wrong, it's merely that I prefer to look at the situation differently.

The main difference in my perspective, which has been my perspective for years now, and posted on these forums over the years, is that when running multiple VGAs with high resolutions, it's not raw CPU core speed that matters...it's memory bandwidth.

So, with that in mind, any perspective that explores performance differences, mainly focusing on raw core performance, or the multi-threaded nature of applications, isn't going to portray the same perspective that I have.


You are right..core perforamnce makes little difference. IN fact, although many sites do not explore this, I feel that the real differences between CPUs in gaming isn't based on mathematical poweress, but is actually more focused on CACHE performance. Of course, because the heirachy of cache design within an OEMs product lines only differs slightly, the actual appearance of core perforamnce in gaming only differs slightly.

I agree 2 FPS is minimal, and unimportant.

But take a look here:




What's going on here? we have a much "slower" CPU, AMD's APU line, giving nearly 50% more performance than the high-end 1100T. Of course, these are not INtel results, but the same plays true. In system configurations that are memory bottle-necked, the 3960X excels.

Perhaps Core performance in gaming, and explorations of such, is a wasted task? If you want to present a certain perspective, yes, it works. But rather than focus on a single aspect, I like to look at hte whole picture, and the whole picture, including my daily use of a 3960X, quite accurately gives me more performance than any other platform does...


And you are right. The extra cores are NOT what gives that performance.


:rockout:
Posted on Reply
#45
radrok
So I guess I was right about the L3 Cache and PCIe lanes, didn't think about the bandwidth on SB-E since SB already has a massive bandwidth with "only" dual channel :)
Also Dave do you know why many bench suites do not see the additional bandwidth quad channel should yield?
Posted on Reply
#46
cadaveca
My name is Dave
Yeah, I'm not saying that every application is going to be faster in a large way. But usage scenarios that do take advantage of the extra bandwidth that the platform provides, do excel. There are beenfits to be had, but I connot deny that the average user isn't going to benefit from it. Those that run 3 or 4 VGAs, and multiple monitors...yes, they will.

I mean really, 1155 is Intel's entry-level platform, IGP included. It is a killer entry level paltform though, without a doubt, and will provide 99% of users will more power than they need. I am not one of those users.

The extra power consumption is a tough pill to swallow, but like I care. For the games I play, it's better. :laugh:

SiSoft Sandra is the only bench right now that shows the extra Bandwidth. It's just about the algorithms used for the tests that are at fault. Many need updates, but I do use Sandra in my memory reviews, so you can see the difference there, at least.
Posted on Reply
#47
radrok
I have my x79 build on hold because I wasn't sure about the performance gains after taking a look at the creativity section on various reviews, I think I'd better wait for the full 8/16.
The only thing that makes me puzzled is that SB-EP Xeons won't OC because they don't have the BCLK untied.
What do you think? I mostly use V-Ray and it's a thread swallower (literally :D)
Posted on Reply
#48
cadaveca
My name is Dave
Really, if you want affordable highly threaded, FX-8150 is the CPU of my own choice. Cost/Perforamnce is damn good.

If it must be X79, 3930K with and OC warranty, with 4.8 GHz is damn snappy. Xeons, with no OC, and no OC warranty, are not something I'd personally consider.


For mutli-GPU gamers on X79, the 4-core with HT, if priced right, would be the CPU to get, because as mentioned the extra cores don't offer much, overall.
Posted on Reply
#49
claylomax
qubitTPU is a fantastic forum and you'll get out of it if you don't put the effort to treat people right and stay within the forum rules
Fixed it for you! :D
Posted on Reply
#50
PaulieG
cadavecaYeah, I'm not saying that every application is going to be faster in a large way. But usage scenarios that do take advantage of the extra bandwidth that the platform provides, do excel. There are beenfits to be had, but I connot deny that the average user isn't going to benefit from it. Those that run 3 or 4 VGAs, and multiple monitors...yes, they will.

I mean really, 1155 is Intel's entry-level platform, IGP included. It is a killer entry level paltform though, without a doubt, and will provide 99% of users will more power than they need. I am not one of those users.

The extra power consumption is a tough pill to swallow, but like I care. For the games I play, it's better. :laugh:

SiSoft Sandra is the only bench right now that shows the extra Bandwidth. It's just about the algorithms used for the tests that are at fault. Many need updates, but I do use Sandra in my memory reviews, so you can see the difference there, at least.
cadavecaReally, if you want affordable highly threaded, FX-8150 is the CPU of my own choice. Cost/Perforamnce is damn good.

If it must be X79, 3930K with and OC warranty, with 4.8 GHz is damn snappy. Xeons, with no OC, and no OC warranty, are not something I'd personally consider.


For mutli-GPU gamers on X79, the 4-core with HT, if priced right, would be the CPU to get, because as mentioned the extra cores don't offer much, overall.
So, Dave, what is the benefit of x79 vs P67/Z68 besides the extra bandwith? Beyond gaming with multi gpus, what usage scenarios would justify choosing x79 over the significantly cheaper platform? I've tried to come up with a shred of justification to do the upgrade, just like I do with every new platform. However, finding this justification has been more difficult than usual. Really not being rhetorical here...
Posted on Reply
Add your own comment
Apr 16th, 2024 09:02 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts