Puget Systems print logo
Read this article at https://www.pugetsystems.com/guides/675
William George (Product Development)

Actual CPU Speeds - What You See Is Not Always What You Get

Written on July 9, 2015 by William George

Intel LogoClock speeds on modern Intel processors are not straight-forward, a fact that is attested to by the several articles we have published on that topic in recent years. This can lead to confusion over what CPU to pick when configuring a new computer, especially for higher-end workstations and servers with high core count processors. I am hoping this blog post will shed a little light on that subject, and help readers be better able to select the right CPU for their needs.

First off, some background. All modern Intel processors have multiple cores, anywhere from 2 to 18 per physical CPU in the models we carry. In addition to listing the number of cores, Intel also advertises the base clock speed of each CPU they manufacture. However, that number is the worst-case speed: it is the speed the processor is guaranteed to run at *at a minimum* when under load. The CPU can drop below that to save power when idling, but under a steady load it will run at least at the base clock speed.

Most of Intel’s processors these days have a feature called Turbo Boost, though, and this is where things get tricky. Turbo boost will increase the clock speed of a processor when under load, to boost performance. The amount it goes up depends on how many of the total number of cores are active, as well as temperature; if the CPU gets too hot, it will back off toward the base clock speed to keep from overheating. In a well-built and well-cooled system, though, that should never happen - so the practical clock speed numbers to look at when specing a computer out are actually the max Turbo Boost speeds.

Intel provides the max single-core boost speed for each processor on their Ark pages, but that can be a misleading number to look at as well. It is quite accurate, as our testing here on systems has shown, but it is for when only a single core on the processor is active. Some applications utilize only a single core, so for those situations it is a good metric, but many use either some or all of the cores a CPU has available. Further, it is rare these days for someone to have just a single application running on their computer; multitasking is the norm, not a rarity. Because of that, the maximum Turbo Boost speed when all cores are being used is also helpful… but that info is buried in Intel white papers. Our Labs department has done a great job collating a lot of that info, though, which I present here:

[-] Hide for Easier Reading
Processor Core Count Base Clock All-core Boost Maximum Boost
Core i7 6700K 4 4.0GHz 4.0GHz 4.2GHz
Core i7 6800K 6 3.4GHz 3.5GHz 3.8GHz
Core i7 6850K 6 3.6GHz 3.7GHz 4.0GHz
Core i7 6900X 8 3.2GHz 3.5GHz 4.0GHz
Core i7 6950X 10 3.0GHz 3.4GHz 4.0GHz
Xeon E5-1620 V4 4 3.5GHz 3.6GHz 3.8GHz
Xeon E5-1630 V4 4 3.7GHz 3.8GHz 4.0GHz
Xeon E5-1650 V4 6 3.6GHz 3.8GHz 4.0GHz
Xeon E5-1660 V4 8 3.2GHz 3.4GHz 3.8GHz
Xeon E5-1680 V4 8 3.4GHz 3.6GHz 4.0GHz
Xeon E5-2603 V4 6 1.7GHz - -
Xeon E5-2609 V4 8 1.7GHz - -
Xeon E5-2620 V4 8 2.1GHz 2.3GHz 3.0GHz
Xeon E5-2623 V4 4 2.6GHz 2.8GHz 3.2GHz
Xeon E5-2630 V4 10 2.2GHz 2.4GHz 3.1GHz
Xeon E5-2637 V4 4 3.5GHz 3.6GHz 3.7GHz
Xeon E5-2640 V4 10 2.4GHz 2.6GHz 3.4GHz
Xeon E5-2643 V4 6 3.4GHz 3.6GHz 3.7GHz
Xeon E5-2650 V4 12 2.2GHz 2.5GHz 2.9GHz
Xeon E5-2660 V4 14 2.0GHz 2.4GHz 3.2GHz
Xeon E5-2667 V4 8 3.2GHz 3.5GHz 3.6GHz
Xeon E5-2680 V4 14 2.4GHz 2.9GHz 3.0GHz
Xeon E5-2683 V4 16 2.1GHz 2.6GHz 3.0GHz
Xeon E5-2687W V4 12 3.0GHz 3.2GHz 3.5GHz
Xeon E5-2690 V4 14 2.6GHz 3.2GHz 3.5GHz
Xeon E5-2695 V4 18 2.1GHz 2.6GHz 3.3GHz
Xeon E5-2697 V4 18 2.3GHz 2.8GHz 3.6GHz
Xeon E5-2697A V4 16 2.6GHz 3.1GHz 3.6GHz
Xeon E5-2698 V4 20 2.2GHz 2.7GHz 3.6GHz
Xeon E5-2699 V4 22 2.2GHz 2.8GHz 3.6GHz

[+] Expand CPU Comparison (2015 models)

Turbo BoostNow why does all this matter? Well, once upon a time you had to decide between a higher number of cores or higher clock speed. Some applications scale well across cores, so higher core counts would make sense even if the clock speed per core was lower. Many applications do not scale well, though, and so are limited to a few or even one core that they can actively use. In those cases, increasing clock speed is going to be the way to improve performance. But what if you have applications with both characteristics? Or even a single application which has some portions that are single-threaded and other parts which can use many cores? Prior to Turbo Boost you had to choose which was more important to you - but now you can have your cake and eat it too (so to speak).

For example, lets say there is user who runs the whole gamut of Adobe's Creative Suite. For most of Photoshop, only 1-4 cores are really utilized... but when rendering with Premier Pro, all of the available cores can be active. Assuming a moderate workstation budget, processors like the Xeon E5-1630 V3E5-1650 V3, and E5-1660 V3 would all be in the running. The first has only four cores, which is plenty for Photoshop - and they run at a nice, high clock speed. For Premiere, though, that might be too limiting... so moving up to the six- and eight-core models gets you additional cores. The *base* clock speed looks lower, though, which might worry the user when it comes to Photoshop performance - until you look at the turbo boost speeds. With just one core active the E5-1650 V3 ends up at the same high 3.8GHz clock speed that the E5-1630 can reach, while having two additional cores available for applications that need them. The E5-1660 V3 doesn't reach quite as high, but the 3.5GHz max boost speed is still respectable and more than a 15% increase from the base speed.

Things get even more interesting as you go to higher core count processors, and if you look at applications which will harness all available cores. Things in the scientific and research areas are often like this. That means the max turbo boost is no longer what is important, but instead the all-core turbo boost speed. This is what these processors will run at when under full load, as long as they are being adequately cooled. That means this is the best metric to use when comparing processors for heavily threaded applications, as I will demonstrate.

Lets say there is a user setting up a system for research at a university. The intended application is well threaded and should use all available CPU cores - but is of course also sensitive to clock speed. Here are two Xeon E5 series processors which are very close in price, and being considered for use (as a pair, but we'll look at the individual CPUs here):

Xeon E5-2687W V3  10-core @ 3.1GHz

Xeon E5-2690 V3  12-core @ 2.6GHz

From those numbers, it looks like a dead heat between the latter two options. One has 20% more cores, the other 20% higher base clock speed. How would you choose? Well, if we look instead at the maximum all-core turbo speeds a different picture is painted:

Xeon E5-2687W V3  10-core @ 3.2GHz

Xeon E5-2690 V3  12-core @ 3.1GHz

In reality, with all cores active, the E5-2690 V3 is almost as fast per-core as the E5-2687W V3, even though the latter costs more and has fewer cores. Now the winner is clear, with the E5-2690 V3 being the best option for this usage case. And to hearken back to the previous example, what if this same user has some code that is only single-threaded? Both the CPUs above max-out at 3.5GHz with a single core active, so there is no downside to the choice even in that scenario.

Hopefully the table and examples above are helpful to those looking to configure a high performance computer! The data will be outdated when the next hardware generation comes out, of course, but the principles should remain sound as long as Intel has Turbo Boost (or some similar technology) integrated into their processors.

If you are looking to purchase a new computer system, please contact our sales consultants! We would love to help you get just the right system for your needs. You can also use the links on the right-hand side to view and configure some of our most popular workstations.

Tags: Intel, CPU, Core, i7, Xeon, cores, clock, speed, turbo, boost

Very very helpful! Thank you. Been looking for this all over the place. So many people just look at Turbo and assume all cores are that speed, I knew it couldn't be true.

Posted on 2015-08-02 02:54:04

I don't think we ever compiled comprehensive lists like this for the later Xeon Scalable processors. However, for any of the models that we carried we do have product pages with specs - including all three clock speeds shown here (base, all-core turbo, and max turbo). Here are some examples:



Posted on 2020-11-10 18:53:59

I just upgraded from a Core2quad to dual Xeon E5 2630 v3 CPUs.
RAM increased from 8gb to 128gb.
GPU went from GTX680 to Titan X 12gb.

For the life of me, I can not get 4k video to play without dropping a lot of frames on the new hardware. To my disbelief, CPU use is only 3% use in Taskman. This on Win 7 64-bit. I changed power profile to performance, but it did't help.

Posted on 2015-08-08 02:13:21

Since the hardware you put in your pc should run 4k just fine I assume that you have made some other mistake..

If you are using DVI or VGA (D-Sub) or even HDMI, 4k just wont happen at acceptable frame rates (DVI would give around 17frames for example due to cable limitations & HDMI might give you up to 30 frames max).

Since you are using a Titan X video card, I advice you to use the display ports to connect it to your monitor, this is the only thing that supports 4k (at a decent frame rate). also make sure you use a display port 1.2 cable, a 1.0 will only run a resolution of 2560×1600 (which is about half of 4k).

hope this helps.

Posted on 2016-01-13 09:37:52

Hi, Many thanks! Do you have the table for older processors?

Posted on 2015-10-09 08:35:00

No, we don't - I'm sorry. You can look up base and maximum single-core speeds on Intel's ARK website, but they don't list max all-core turbo speeds: http://ark.intel.com/

Posted on 2015-10-09 15:20:42

I am trying to decide between the 4th gen i5-4690 vs i5 4590T. Both have 6 MB cache but the i5-4690 has 3.5 GHz base and 3.9 GHz max turbo while the i5-4590T has only 2 GHz bas and 3 GHz max turbo. Both are 4 cores and 4 threads. Would I notice much of a difference in speed?

Posted on 2016-02-26 17:31:48

That is a pretty big difference in clock speed - almost 1GHz at max turbo, or ~30%. I'd say that will likely be noticeable, especially if you do anything demanding with the system. Just internet / basic usage would likely be fine on either.

Posted on 2016-02-26 17:44:29

ok thank you!

Posted on 2016-02-26 17:50:24

Hi, thanks for the informative article. I have compared the Intel Core i7-4750HQ @ 2.00GHz to the Intel Core i7-4720HQ @ 2.60GHz:ore i7-4750HQ @ 2.00G
https://www.cpubenchmark.ne...[]=1957&cmp[]=2448. Why does the first (with only 2.00 GHz) get the best rating?

Posted on 2016-03-09 16:30:05

I try again with the link:

Posted on 2016-03-09 16:34:38

Those are laptop CPUs, and while the model numbers are very similar they are different under the hood. The 4750HQ is older, has lower clock speed, is a soldered-on type CPU, and has better built-in graphics - while the 4720HQ is newer, higher clock speed, and a socketed CPU. If you look at the PassMark details for each CPU, the 4720HQ actually has a higher single-threaded score... likely due to the higher clock speed. I'd guess that the 4750HQ pulls ahead (just slightly) in overall benchmark scores because of its better built-in graphics, but without seeing all the details of the test results (which they don't display on that website) I can't say for sure.

Posted on 2016-03-09 16:49:57

Thanks for your answer. Which one should I prefer when buying a new laptop? The price differs with $70.

Posted on 2016-03-09 17:04:47

Honestly, those are both a little on the older side... do you have any options for a more modern CPU? Not that they'd be bad, but they are 1-2 generations old (depending on how you count it) and use older memory as well.

If those are your best options, look at the graphics setup. If the system has a dedicated video card, then the 4720HQ with faster clock speed will be better - but if they use built-in / Intel graphics, then the 4750HQ's better graphics will probably outweigh its slower clock speed.

Posted on 2016-03-09 17:12:41
Christo Cordier

Good day , please help me out ... Xeon E5-2690 V3 12-core , i find two variants one with a 2.6ghz base clock and another with 2.4 ghz base clock (ebay) what is the difference , both supposed to be the V3 , can they both work in X99 motherboards? (mainly msi X99A sli motherboard)
thank you for any help .

Posted on 2016-04-18 00:38:59

Hmm, I don't see any record of a E5-2690 (from any generation: v1, v2, v3, or the brand-new v4) that runs at a 2.4GHz base clock speed. Maybe it is a typo? Either in the speed or the model number? Be careful where you order from, though, that it is a reliable source.

As for working on a X99 motherboard, I cannot speak for MSI. We've used several Asus X99 boards, and all seemed to work fine with the Xeon E5 v3 processors. We also used one ASRock, which again worked fine. If you haven't purchased anything yet, though, you might want to wait a little longer for the new v4 processors to become more widely available - and for motherboards to get BIOS updates to support them.

Posted on 2016-04-18 04:34:28
Christo Cordier

Thank you very much.
According to the (eBay) it's an engineering sample... I'll rather pass on that atm and wait for the v4 as you suggested. Thanks again.

Posted on 2016-04-18 18:12:45
Tom Myles

Very interesting, do you know if there is similar data for the current v4 processors? I seem to be going around in circles on the Intel site without finding anything useful.

Posted on 2016-06-08 13:07:29

You are right - Intel makes it very difficult to find some of the turbo speeds! Our Labs folks have done a good job digging, though, and for all the processors where we have found that info we now list it on the product specs pages on our website. We've also started to put those numbers in the name of the processors, for easy comparison on our configure pages. If you go, for example, to the Desktops -> Custom Computers page and select one of the X99 motherboards you will see the Core i7 and Xeon E5 processors listed - and they should have, in the name, the base speed along with the turbo boost range (from max all core to max single core).

Posted on 2016-06-08 15:36:23
Joe Johnny

Interesting.. My 4790k hits 4.4 on all cores at once and over 4.5 without any overclock (not sure if all at once). Using a Gigabyte Z97 Gaming 3 motherboard and Hyper 212.

Posted on 2016-07-02 05:20:40
Jake Gabriel

Hey William,

I've got a couple questions for you. Currently I'm running two 2620 v3's (not m first choice but got them for a steal) with an asus Z10PE-D16 WS. I'm currently using them for x264 encoding and they are doing very well. With this motherboard I've pinned the CPUs at top all core turbo and turned the base clock to 103.0, putting the constant clock speed at about 2.68 GHz on all cores.

My questions are,

Am I hurting my CPUs?

If I stop pinning the CPUs at max all core turbo will I get better performance with x264 since it only uses 18 or so threads? I wanna squeeze the most performance out of these chips as I can.

I would greatly appreciate any response!


Posted on 2016-11-18 06:37:31

I haven't played with changing clock speed settings on these dual Xeons, honestly, but the 2620 v3 should run at 2.6GHz when all cores are under load... so 2.68GHz isn't much faster, and probably not worth the trouble. However, it is also probably not high enough to be causing any damage to the CPUs, if they are well cooled.

The question I would have is whether you no longer get the single-core boost to 3.2GHz? If that is not happening, then you probably are getting lower performance overall. Many applications will not use all the cores, and even applications which are well threaded often have some parts which don't use more than one (or a couple) cores.

Posted on 2016-11-18 18:06:20

Hi, I have a E5-2690 v3 in a Asrock x99 Fatal1ty Professional Gaming Mobo. Everything works fine except that it show 2.4ghz in the bios screen instead of the 2.6ghz that it is supposed to be. I thought it should definitely show 2.6 since I have done no tweaking on the board at all. thanks.

Posted on 2016-12-12 03:27:52
mike dar

I'm curious if, with Turbo in on, would Hyper-Speed stack performance another level up and be possible with these Xeon E5-2699 V4 chips.
I understand the BLK is all that can be changed with the feature(Hyper-speed) I believe and that the 'extra' with 'hyper' is about 5-6% 'faster' speeds.

Posted on 2016-12-19 13:23:23

I am not familiar with the term "Hyper-Speed" related to Intel CPUs. They have Turbo Boost and Hyper Threading, but those don't really impact eachother directly.

I did some Googling, and it looks like SuperMicro (a manufacturer of motherboards and rackmount chassis) uses that term - though I cannot tell for sure what they are implying with it. It sounds in some places like it has to do with ensuring the CPU(s) get enough power that they can stay up at full Turbo Boost speeds... and yet it seems also to have something to do with the memory. Memory controllers are on the CPU, so maybe that is somehow related? Anyhow, it seems to be some extra BIOS stuff that SuperMicro does - and has given a slick name to. I'm sorry that I cannot comment on it more, but insofar as I can tell it doesn't claim to increase performance above what Turbo Boost is designed to provide.

Posted on 2016-12-19 16:31:56
mike dar

As I understand, the 'Hyper' in in the Bios for the X10 boards (2.0 or higher updated bios) where the requirement I believe is for 'Turbo' to be turned on at all times, then the BLK is over volted with as many as 4 possible steps up in voltage only in BLK (What I can Find is that the third step is about the highest stabile for most users) and the bios clocks the ram (only one stick seems to be recommended, I can find from Samsung) to match in higher freq.

I'll look closer, just was curious if anyone had tried one of the Supermicro x10s and Xeon chips with that 'Hyper to see how far it pushed freq;s.

Posted on 2016-12-19 19:41:29

I don't think we've done anything with that. It sounds like a form of overclocking (of the memory, along with overvolting the CPU) which we stay away from unless customers order a system with it - and currently none of the combinations we offer overclocking on include a Supermicro motherboard. Maybe someone else reading this has tried it, though? I'm personally skeptical about it being necessary, as we see full turbo boost speeds maintained without issue on the systems we build with Asus motherboards.

Posted on 2016-12-19 20:13:09
Niko Nikolov

This is a incredible piece of info.In this moment i am facing a problem which i think this information is solving.In my case i need a ¨¨all in one cpu¨¨

So im doing Gaming,unreal engine work(4 cpus maks used there),photoshop,video editing,cpu rendering with keyshot,autodes maya heavy polygone modeling and even agisoft photoscan.So in my case more cores and good speed is required.That leaves me with e5-2690 v4 vs e5-2687w v4.Their base clock is wayyy different,but if I understood this right,I have to pay atention only to the all core turbo which is the same.And if im not mistaking, the base clock of 2.6 ghz vs the 3ghz,this can only happen if thermal throttling to that optimum speed if case ventilation or cpu cooler is not good.But with a good noctua cooler and proper ventilation(and depending on the application)those scarry 2,6 shoud be 3.1-3.2 ghz on all cores at sustained time of use/days? In this case the e5-2690 v4 shoud be the same performer as the e5-2687w v4 in speed but cheaper and less w tdp. Did i get this right?

Posted on 2016-12-31 16:09:40
Niko Nikolov

This is a incredible piece of info.In this moment i am facing a problem which i think this information is solving.In my case i need a ¨¨all in one cpu¨¨

So im doing Gaming,unreal engine work(4 cpus maks used there),photoshop,video editing,cpu rendering with keyshot,autodes maya heavy polygone modeling and even agisoft photoscan.So in my case more cores and good speed is required.That leaves me with e5-2690 v4 vs e5-2687w v4.Their base clock is wayyy different,but if I understood this right,I have to pay atention only to the all core turbo which is the same.And if im not mistaking, the base clock of 2.6 ghz vs the 3ghz,this can only happen if thermal throttling to that optimum speed if case ventilation or cpu cooler is not good.But with a good noctua cooler and proper ventilation(and depending on the application)those scarry 2,6 shoud be 3.1-3.2 ghz on all cores at sustained time of use/days? In this case the e5-2690 v4 shoud be the same performer as the e5-2687w v4 in speed but cheaper and less w tdp. Did i get this right?

Posted on 2016-12-31 13:57:03

Yes sir, that has been our experience in testing :)

Posted on 2017-01-02 18:47:54
Niko Nikolov

Awesome!I actually feel better hearing it from people who do this,rather than hunting intel devs on their forums xd.

Posted on 2017-01-02 19:18:20

So in other words, how do I get my Xeon E5-2670v3 to go past the 2.6Ghz All boost, to the 3.1Ghz Maximum boost? Do I actually have to turn hyperthreading off?
My situation is when I open a software that uses all my cores, the max speed at full load I get goes from the 2.3Ghz to (technically) 2.58Ghz and then stops at that level and continues with the processing at 2.58Ghz for the remainder of the computational process. I bought the chip because it says TurboBoost 2.0 up to 3.1Ghz. How do I achieve that speed? I have everything enabled in the BIOS.
I think I know the answer to my question but I just want to hear thoughts from others. Thanks!

Posted on 2017-02-08 15:21:45

If your software is using all of the cores then you're only going to see the all-core boost speed. The maximum turbo boost is only reached when a single core is active and the others are all idle.

Posted on 2017-02-08 17:12:53
Mayank Magoo

Very useful info. Could you add 2699A V4 to the above list. I am interested in it's all core turbo frequency.

Posted on 2017-03-09 05:19:05

Hmm, it looks like that CPU variant came out later than most of the other E5 v4 processors. It is not mentioned in the Intel documentation we have, which is where we got the all-core turbo speeds for the other processors (which we have also verified in our own testing). We don't carry / use the 2699A v4 either, so I don't have any internal data from testing to look at.

What I do see is that the core count is the same as the normal 2699 (non-A), and the max single-core turbo speed is the same (3.6GHz)... but the base clock speed starts off a little higher (2.4 instead of 2.2GHz). If I had to guess, the all-core turbo could be as much as the same 0.2GHz higher - which would make it 3.0GHz - or it could be the same as the non-A model at 2.8GHz. That probably doesn't help much, but I don't have any solid answer for you. Without solid info, I cannot add it to the list above either.

Posted on 2017-03-09 18:44:29
Beta Zoid

Hi William, can dual e5-2660 v4 be used for vfx work in single threaded applications, such as 3ds Max, Maya and caching processes with a 2.0ghz base clock?

Posted on 2018-01-17 12:23:51

Great article! Wondering if William M George could advise me on a configuration dilemma. Need to know which system would be better for video editing, rendering, graphics, etc. using Adobe CC and some insight as to why. Thank you in advance for your advice...much appreciated!

System 1: Dual Xeon e5-2623 v4 2.6GHz with 32GB of DDR4 2133 RAM and NVIDIA Quadro M2000 4GB video (Dell Precision Tower 7810)
System 2: i7 6700K 4.0GHz with 32GB of DDR3 1867 SDRAM and AMD Radeon R9 M395X with 4GB video (iMac)


Posted on 2017-04-15 00:02:54

Hmm, honestly neither of those is ideal. Assuming you are using Premiere Pro, you'd be better off with more than the four CPU cores that System 2 has - but you also want to keep the clock speed up, and System 1 is severely lacking there. If you are stuck with only these options, I'd say System 1 if you tend to focus on one application at a time... or System 2 if you multitask and have a lot of programs open and do things like rendering in the background while working in a different application.

But better yet would be a system with a single 8-10 core processor at high (3GHz+) clock speeds and a newer NVIDIA graphics card. We've done a lot of specific testing on Adobe programs and have recommended computers for these applications:


We've also done testing on MacPro vs PC systems, and dual CPUs vs single:



Those, and many of our other articles, would probably interest you :)

Posted on 2017-04-15 00:12:20

Hello. I really appreciate the ultra-fast response...much appreciated! I'm definitely multitask and I'm usually running multiple Adobe CC programs in my workflow with other office programs open at the same time as well. These are my only two options (workstation at work) so I need to choose between the two. I did actually find you from searching and finding other Puget System articles as well such as:


After sending this article to my IT dept, they suggested disabling multi-threading on the Dell system (System 1) which they did but wondering if this is even a good idea? I did run my own PP rendering test on a 1 min, 4K timeline without any effects and both times it took about 2 min to finish encoding inside of PP (with multi-threading turned on and off). I was concerned about the overall clock speed of the dual system but assumed, incorrectly, that a server-grade dual CPU would beat out an i7-6700K for the type of work I do. Given this additional info, the fact that I multitask with Adobe PS, AI, and ID on a regular basis, and these two systems are my only choices, is the iMac the hands down winner? @WilliamMGeorge

Posted on 2017-04-15 00:31:58

I don't think turning 'multi-threading' (I'm assuming they mean Hyper Threading) off will help Adobe programs. There are some other situations where that can be a factor, but I don't recall ever seeing it come up with Adobe - so I'd leave that on.

What you really have here is 8 cores at 2.8GHz (all core turbo) vs 4 cores at 4GHz. That is a big enough clock speed delta that I think for individual usage of Adobe programs the Mac with 4 cores will do better... but the trick comes in with your multitasking. If you have a video rendering in the background, the 8 core system should remain more usable than the 4 core would - and foreground applications will have less impact on the rendering performance as well.

You've got other complications too, though: OSX vs Windows, NVIDIA vs AMD graphics... and I didn't see any mention of the drive setup on the two systems. All three of those things could come into play. Given the wide disparity, I think your best bet would be to try both - if that is an option for you - and see which you like better.

Posted on 2017-04-15 00:41:03

I really appreciate the insight...OK, if I ask one final question? So, my video work isn't heavy-duty...just cutting up short 1-2 minute clips for use on social so most of what I would encode won't take but a few minutes. I won't be working on huge timelines with lots of LUTS or other effects. My multitasking in Adobe CC really consists of creating a quick graphic in PS or AI when I'm cutting up video. My bigger concern is the fact that I've already seen this dual Xeon system take 8-10 seconds to save out a multi-layered PS file and I would think the i7 would crunch this file without issue given its substantially higher clock speed, even if PR and AI were running in the background. Unfortunately, I can't try out the other system so I'm stuck trying to get online advice from experts like yourself before making the switch. I understand it's difficult to compare the two because it's not apples-to-apples but I do appreciate your opinions. Thank you! William M George

Posted on 2017-04-15 00:54:31

Hmm, if your work is shorter stuff mostly then you aren't likely to sit there with a render going for minutes or hours at a time in the background - so even if you have many programs running at the same time (which requires a decent amount of RAM) you are probably right that the Mac's faster clock speed would win out for your normal usage. It is also new enough that it ought to have a solid-state drive - both of them would, I'd hope!

Now if it were me, I'd likely still stick with the dual Xeon Windows box just because of the OS: I don't have a lot of experience using OSX. If you are comfortable with it, though, and as these are your only choices... I suppose that is the way to go. Maybe next time you are due for a computer upgrade you can convince your IT staff to get you a system more purpose-built to your needs :)

Posted on 2017-04-15 03:59:49
Beta Zoid

Which Cpu is better and faster. The E5-2697 v2 or E5-2660 v4?

Posted on 2018-01-12 22:33:43

Hmm, that is tough. The 2660 v4 is two generations newer, and uses newer / faster memory as well, but technically the clock speed is about 10% lower (both for all core and max single-core turbo). It also has two more cores, though, and uses less power. I'd say overall that the 2660 v4 is probably the better CPU, and likely to be a little bit faster under multi-threaded workloads. In some single- or lightly-threaded situations they may be more equal, with the newer architecture and faster memory being offset by the lower clock speeds. Its hard to say precisely, though, without testing them in your specific application.

Posted on 2018-01-15 20:17:49
Beta Zoid


Posted on 2018-01-16 01:51:07
Beta Zoid


Posted on 2018-01-16 01:54:24
Beta Zoid

Hi William, can dual e5-2660 v4 be used for vfx work in single threaded applications, such as 3ds Max, Maya and caching processes with a 2.0 base clock?

Posted on 2018-01-17 00:24:59

*Can* it be used for that? Sure... but it won't be as effective / fast as a processor with a higher single-core clock speed. The modern Core i9 processors, for example, will give much faster performance with single- or lightly-threaded applications, while still providing a lot of cores (up to 18 on the i9 7980XE) when needed. They can't be used in dual CPU setups, though - but they will also be less expensive, because of that.

Posted on 2018-01-17 22:32:53
Beta Zoid

Thank you William :)

Posted on 2018-01-18 09:05:50

thank you!. it is very helpfull!.

Posted on 2017-05-28 17:44:20

please help me out ... Xeon E5-2650 V3 10-core 2.3 ghz max turbo frecuency 3.00 GHZ. I set up my system in single cpu with z10pe d16ws . i havent had the max frecuency (3.00 Ghz) when render in 3ds max.max frecuency in oc mode up to 2.72 Ghz.
is there any way to overclock the cpu or the only way for mor performance is that add another cpu on motherboard.

Posted on 2017-05-28 17:55:30
Chris MacDonald

Have only just seen this. I thought I had a dodgy CPU when I noticed my (brand new) dual xeon e5 2630 v4 was rendering @2.4GHz rather than the listed 3.1GHz. Quite dissapointing as the Intel documentation does not make this clear at all.

Is it at all possible to force the max turbo boost on all cores (obviously provided there is sufficient cooling)?

Posted on 2017-07-18 08:08:46

Interesting but in reality while working with Ae ,not rendering so you don't hit max, it will benefit from the higher speed processor? So in this case the
Xeon E5-2687W V3 would win?

Posted on 2020-03-03 13:56:17

Keep in mind that this post is almost 5 years old now, so the CPUs listed are not what you would want in a new system. Still perfectly relevant for an upgrade of an older system, but just be aware of that.

For After Effects, you pretty much want the best single core performance you can get - that means the highest maximum Turbo speeds. Straight clock speed doesn't translate between Intel/AMD or even between product families from each brand, however, so you really need to try to find benchmark results for the exact CPU you are looking at. We don't test anything as old as the 2687W V3 anymore, but this is the latest testing we have available: https://www.pugetsystems.co...

Posted on 2020-03-03 16:18:44

Most applications outside of rendering and some very specific other use cases won't utilize all the cores in a processor, so higher clock speed CPUs will tend to win out as long as they have "enough" cores for the given application. However, the point of this article was that the advertised clock speeds on processors no longer match what speed they actually run at when under load thanks to features like Turbo Boost. This is why real-world testing (like we do here in our Labs department) is really helpful: it can cut through the confusion over how many cores you need, what speed different chips run at, and even how families and brands of processors differ and simply get to the actual performance they provide in common applications that folks run :)

Posted on 2020-03-03 16:46:36

Thanks for the explanation. This actually made me think about the way you compared the two processors.

In all cores situation, wouldn't it make much scene to measure both the cores and speed of the processor? like using Cinebench?

Posted on 2020-03-03 17:16:39

Yes, for applications where all of the cores are going to be utilized fully a benchmark test like Cinebench (which is just rendering, a very well-threaded type of workload) is a good way to compare the maximum / ideal performance of different processors.

Posted on 2020-03-03 17:30:31

Thanks for the help.

Posted on 2020-03-03 17:37:13

In this case of the E5-2687W V3 vs E5-2690v3, if the main usage is virtualization (VMware, Hyper-V, XCP-ng, etc.), the E5-2690v3 would be a better choice? Since "more" cores are essential in virtualization?

Great article! I know this article is old but the information is still relevant in my case.

Posted on 2020-04-26 02:50:55

Most likely yes, but it could depend on what you are doing with virtualization: how many VMs are you running, what is going on inside those VMs, etc.

If your application requires a "real" core to be assigned to each VM, then the 2960v3 would give you two more cores - so you could run two more instances. The clock speed per core when under turbo boost is so close between those two that it would be hard to come up with a situation where the *slightly* higher clock speed on the 2687W would be more important, but there might be a handful of such niche cases.

Posted on 2020-04-27 21:25:50

Have you got a table like this for the next 2 generations for these Xeons - i.e. Xeon Gold 61xx and 62xx?

Posted on 2020-11-10 10:29:23

Hi William, great article!

I am running dual E5-2690 V4’s in a Z840 and I can’t for the life of me get them to run at the all core turbo speed of 3.2 whilst rendering (either in Arnold or when benching in Cinebench). They’ll happily idle at 3.2 when in high performance mode but once I start rendering they immediately drop to 2.89 and stay fixed there. In between frames they jump back up to 3.2 and then as soon as the rendering starts again it’s back to 2.89. Is this how dual 2690 V4’s should behave?

Thanks in advance

Posted on 2021-06-25 15:02:55

I haven't used a pair of those in a long time, but I would suspect that something else might be causing them to throttle down a bit. It could be temperature, power draw, or even just more conservative default settings in the BIOS on your system. I am especially suspicious of that because when only a single core is loaded down (or when they are all idle) those processors should reach 3.5GHz... but you are reporting 300MHz lower speeds both at that time and then under all-core load as well.

Posted on 2021-06-25 18:40:29

Thanks for getting back to me so quickly on such an old thread. Really appreciated! Is there any particular direction you could point me towards to try to work out why I’m having this slight throttle or underclock? My temps are below 80 degrees when rendering, and I’ve turned off all processor c states, power saving etc in the bios. I’ve run out of things to try

Posted on 2021-06-25 20:32:51

Another quick update: I’ve just run a stress test (screenshot attached) via cpu-z and all cores under load for an extended period of time are running at 3.2Ghz as they should be. However as soon as I run a render engine such as Arnold or Cinebench it drops to just under 2.9GHz as in the previous screenshots. This is weird right? https://uploads.disquscdn.c...

Posted on 2021-06-25 20:55:06

Oh, okay I bet I know what's going on! When certain advanced CPU instruction sets are used, particularly AVX2 and AVX512, the CPU may need to run at a lower max speed. Intel doesn't do a very good job of documenting this publicly :(

Posted on 2021-06-26 03:07:24

Hmm, I would look in the bios for settings related to turbo speeds and times. These processors supported turbo boost 2.0, which also introduced the option for some of these turbo speeds to only last for a certain amount of time. I don't remember what Intel's defaults for that were, but I know that different motherboard manufacturers implemented it differently. It could technically be construed as overclocking, but if you were able to remove those time limits or lengthen them you might see the system reach the desired clock speeds. Wikipedia has a little more info: https://en.m.wikipedia.org/...

Posted on 2021-06-25 23:17:11