Intel Core i9 Skylake E Clocks no Higher than Broadwell E.

[FrgMstr]

Doing my regular round of interneting this morning looking for good news to pass along, cageymaru passed along this tidbit from Guru 3D that I found very interesting. Guru is passing along some of its findings out of Computex about the upcoming Intel Core i9 CPUs. First and foremost, it reported that the new 10-Core 7900X has been spotted pulling a Cinebench benchmark at 4.3GHz on a Liquid Cooling System (LCS), scoring 2,364. Now, while the 10-C Broadwell 6950X CPU was not that popular at $1400 in retail (7900X is reported to cost $1000), it was still easily pushed into the 4.3GHz range with a simple Corsair H80 cooling system with a push/pull fan setup. I would know because that is exactly what my system is set up with right now. I built this system recently to render 4K video, which it does quite well. Also this system will easily score close to 2,300 in Cinebench with the 2800MHz memory it is running. So all in all, the Skylake-E does not look much different than Broadwell-E, as many of you probably expected.

Also a line that caught my eye is this one in regards to LNG overclocking.

Its getting old and it is not at all important or relative towards you guys, the end user.

No, LNG overclocking has not been important or relative literally since its inception to the end user. Fun? Yes! Entertaining? Yes! But it not "getting old," it has been old for quite a while and Intel's lack of clocking scale in any significant form in its retail products is the reason LNG is shoved down our throats like it is actually meaningful.

Also Guru went on to say this about CPUs available at launch.

None, and I do repeat this none of the partners have had their hands on, or even seen the 12, 14, 16 or that 18-core part. So that does raise some questions as earlier on we have already stated that Intel is rushing things as an answer towards AMD’s upcoming Threadripper processor series. So on that note, we can confirm that X299 and the respective processors will launch with up-to 10-core processors in the Skylake-X processor series. So that means that there will be three (Skylake-X) SKUs at launch, the Core i7 7800K six core, the 7820X with eight cores and the Core i9 7900K with 10 cores.

As suspected, the Intel Core i9 7920X, 7940X, 7960X, and 7980EX seem to be nothing more than a kneejerk reaction to AMD's ThreadRipper, and Intel getting "Xeon" parts out in retail packaging. AMD is truly being disruptive in the market place, and that is a great thing to see. Intel needs to wash its hands now, as it has seemingly pulled its thumb out of its ass.

 

[Jorona]

Pfft. They won't wash their hands. They enjoy the smell of their own shit.

 

[SomeoneElse]

They truely have competition now. Lets see how much innovation comes from this. I for one am glad to see and welcome the fact they AMD is relevant again. I run Intel right now but i'm not a fan boy but always go for the best price and best performance. I ran AMD when they broke through the 1 GHz barrier years ago.....lets see it heat up the market again.

 

[Nukester]

I don't know Kyle, I don't know if they will ever pull their thumb out of their ass when it comes to product features/pricing. That's what happens when a company is king.

 

[Quartz-1]

Threadripper and Epyc aren't out yet. The proof of the pudding will be in the [H] review.

 

[Gigus Fire]

Competition is good and benefits the consumer.
Because intel had practically no competition is why advancements have stagnated and prices have gone up.

 

[Filiprino]

Intel needed 5 Ghz on the i7 7700K to surpass a 4 Ghz 1700... These many-core SKUs won't have that advantage.

 

[Nimisys]

It'll be interesting to see what the launch premium is going to be on those three chips. I wonder how artificially constrained the marketplace will be on those.

 

[Teenyman45]

Despite Intel kneejerk reactions you still need to wonder at some what the company is doing to artificially create market segmentation. I've mentioned before and point out again that the 7800K "entry level" SkylakeX, in addition to a lower core count, has reduced memory support of only 2,400MHz rather than 2,666 like the rest of the product line.

If the Cinebench score is roughly the same then I guess increasing the L2 while cutting L3 cache didn't have much improvement. I'm also curious how the cache change affects temperature and power usage. In the same vein, is the overclocking cap on water due to voltage limitations on the chip or is it thermal throttling as a result of using tim instead of solder?

Threadripper and Epyc aren't out yet. The proof of the pudding will be in the [H] review.

Unless there is a real problem with the infinity fabric or from giving the individual Ryzen blocks additional i/o capabilities from all the extra PCIe lanes, then clockrate IPC should be no different and it will just be a matter of how well the OS scheduling and programs can use the extra cores compared to regular Ryzen.

 

[M76]

If 7900X is $1000 then what will the rest cost? That's as much as most people pay for an entire setup. Not just the CPU. I want 8 core part for under $500 if they hope to compete with amd.

 

[Sikkyu]

If 7900X is $1000 then what will the rest cost? That's as much as most people pay for an entire setup. Not just the CPU. I want 8 core part for under $500 if they hope to compete with amd.

ryzen 1700 is already well under 500. I really want to know what a full power Threadripper will run. Under 1000? I might go for it.

 

[M76]

ryzen 1700 is already well under 500. I really want to know what a full power Threadripper will run. Under 1000? I might go for it.

That'S why intel needs to step up their game. They have no counter for Ryzen yet, and I doubt I9 will be price competitipe with threadripper. And it seems to me they're not trying to compete with the prices, for some weird resaon they assume that enthusiasts pockets are bottomless. Well it's not. And even if it was, I'd still buy the part with the best bang for buck.

 

[GoodBoy]

Intel needs to wash its hands now, as it has seemingly pulled its thumb out of its ass.

Lol!!

Thank you AMD for getting Intel out of its' complacent daze its been in for years...

Now, how long until some actual new and improved cpu's finally see the light of day..

Just starting to read the threadripper article... The fact that all these new AMD cpu's will have 64 pcie lanes is pretty sweet. Intel has been lagging so badly in this regard. The fact that only the 7900x even has 44 lanes, and those other high core cpus have only 28 is bullshit. Seriously a half-assed attempt to compete in the high core cpu arena...

Get off your fucking asses and build something truly competitve Intel...

Going to finish reading the threadripper article, can't wait for [H] to get it's hands on these new processors.

 

[applejacks]

do these new cpus support dual cpu or did they leave that just for the Xeons still? Aren't these new chips basically Xeons
I'm on the fence about putting together a dual e5-2680 v3 setup, just missing the motherboard at this point.

 

[sirmonkey1985]

do these new cpus support dual cpu or did they leave that just for the Xeons still? Aren't these new chips basically Xeons
I'm on the fence about putting together a dual e5-2680 v3 setup, just missing the motherboard at this point.

i can't see either company releasing dual socket boards for those processors.. to much of a risk ruining xeon/opteron sales.

 

[Teenyman45]

i can't see either company releasing dual socket boards for those processors.. to much of a risk ruining xeon/opteron sales.

Threadripper is already like a dual socket with the double eight core chips mounted on the big platform.

 

[The Cobra]

Where is Threadripper? That is the one I am waiting for...

 

[ZLoth]

In September 2014, I upgraded from a Intel Core 2 Duo E6600 to a i7-4790K "Devil's Canyon" based system. And, there was a noticable improvement in speed between the two systems. Now, how much of a speed improvement would I see if I upgrade from a i7-4790k to a current system? Ten percent? Twenty....maybe?

I'm already at a dead end at upgrading the system.

Mo memory? I'm already at 32GB for video editing and virtualization.

Storage? I have a 1TB SSD main drive and a 256GB used SSD that is a scratch drive. As for backups and archive files, I have a FreeNAS server with eight 5TB drives in a RAIDZ2 configuration.

Monitor? I have a 27" G-Sync primary monitor running at 2560x1440 resolution, and I have a second 27" 2560x1440 IPS monitor 60Hz monitor "on it's way" for work purposes. No multi-monitor gaming here.

Video card? I have a 980 video card, and I'm looking to replace it with a 1080Ti at the end of the year. Maybe.

The sad truth is that, unless you are a [H]ardGamer or have some [H]ardApplications, the lower-end processors are (G)oodEnough for web browsing and watching videos.

 

[luke51087]

In September 2014, I upgraded from a Intel Core 2 Duo E6600 to a i7-4790K "Devil's Canyon" based system. And, there was a noticable improvement in speed between the two systems. Now, how much of a speed improvement would I see if I upgrade from a i7-4790k to a current system? Ten percent? Twenty....maybe?

I'm already at a dead end at upgrading the system.

Mo memory? I'm already at 32GB for video editing and virtualization.

Storage? I have a 1TB SSD main drive and a 256GB used SSD that is a scratch drive. As for backups and archive files, I have a FreeNAS server with eight 5TB drives in a RAIDZ2 configuration.

Monitor? I have a 27" G-Sync primary monitor running at 2560x1440 resolution, and I have a second 27" 2560x1440 IPS monitor 60Hz monitor "on it's way" for work purposes. No multi-monitor gaming here.

Video card? I have a 980 video card, and I'm looking to replace it with a 1080Ti at the end of the year. Maybe.

The sad truth is that, unless you are a [H]ardGamer or have some [H]ardApplications, the lower-end processors are (G)oodEnough for web browsing and watching videos.

The improvement you would see would be if you are capping out your 4/8 core/thread cpu. And of course maybe using a new nvme drive. Really does depend on what you are doing but it should come down to price to performance, and right now i dont think Intel has an answer for ryzen.

 

[Oniigumo]

In September 2014, I upgraded from a Intel Core 2 Duo E6600 to a i7-4790K "Devil's Canyon" based system. And, there was a noticable improvement in speed between the two systems. Now, how much of a speed improvement would I see if I upgrade from a i7-4790k to a current system? Ten percent? Twenty....maybe?

That's nearly 8 years difference between the launches of those CPUs, wait till Q1'22 to make the 4790k to *latest CPU here* speed argument. I completely agree the market stagnated over a few gens, but it seems to slowly be getting better with AMD returning to relevancy in the high end. I made nearly the same leap as you, going from the E6600 to a 3770, then a 4690k because I missed the OCing Will likely stay on this CPU for 3 or 4 more years, unless something comes along and makes meaningful improvements (from a gaming perspective, nothing more).

 

[GameLifter]

In September 2014, I upgraded from a Intel Core 2 Duo E6600 to a i7-4790K "Devil's Canyon" based system. And, there was a noticable improvement in speed between the two systems. Now, how much of a speed improvement would I see if I upgrade from a i7-4790k to a current system? Ten percent? Twenty....maybe?

I'm already at a dead end at upgrading the system.

Mo memory? I'm already at 32GB for video editing and virtualization.

Storage? I have a 1TB SSD main drive and a 256GB used SSD that is a scratch drive. As for backups and archive files, I have a FreeNAS server with eight 5TB drives in a RAIDZ2 configuration.

Monitor? I have a 27" G-Sync primary monitor running at 2560x1440 resolution, and I have a second 27" 2560x1440 IPS monitor 60Hz monitor "on it's way" for work purposes. No multi-monitor gaming here.

Video card? I have a 980 video card, and I'm looking to replace it with a 1080Ti at the end of the year. Maybe.

The sad truth is that, unless you are a [H]ardGamer or have some [H]ardApplications, the lower-end processors are (G)oodEnough for web browsing and watching videos.

I built a PC with a 3770K in 2012 that I am still using and I feel no need to upgrade the CPU just yet. All I've had to upgrade over the years was the GPU since I mainly use my PC to game. I've thought about waiting for Coffee Lake to make a decision on what CPU to put in my next build but I may wait even longer than that. I'm thinking about putting my money that I would have spent on a new build towards one of the new 4K/144Hz GSYNC HDR monitors. With this upgrade I would notice a clear difference and the bottleneck would once again be more on the GPU so if I want extra performance I could simply upgrade my GPU.

 

[Domingo]

As a gamer, there's no reason for me not to stick with my 4790K indefinitely. The biggest motivator would (sadly) be the ability to play 4K videos with DRM, although I guess that's coming to 9x and later Nvidia cards this fall anyway.

 

[Gigantopithecus]

In September 2014, I upgraded from a Intel Core 2 Duo E6600 to a i7-4790K "Devil's Canyon" based system. And, there was a noticable improvement in speed between the two systems. Now, how much of a speed improvement would I see if I upgrade from a i7-4790k to a current system? Ten percent? Twenty....maybe?

The sad truth is that, unless you are a [H]ardGamer or have some [H]ardApplications, the lower-end processors are (G)oodEnough for web browsing and watching videos.

I just upgraded from a 4790K (OC'd to 4.4GHz) to a 1700X (OC'd to 3.8GHz), and you're right, for basic day-to-day functions, they're indistinguishable. But do you really consider video encoding and photo editing [H]ard? Because with those two tasks, the 1700X blows the 4790K out of the water. And the 1700X has nearly doubled my ability to do bona fide [H]ardApplications required by my scientific research.

Doubling core and thread counts has never made all that much difference for basic stuff - hell, going from my Athlon 64 3000+ to Athlon 64 X2 3800+ only helped with audio and photo editing (I wasn't editing videos back then), and my research applications. It didn't help me type papers or make PowerPoints.

 

[egads]

In September 2014, I upgraded from a Intel Core 2 Duo E6600 to a i7-4790K "Devil's Canyon" based system. And, there was a noticable improvement in speed between the two systems. Now, how much of a speed improvement would I see if I upgrade from a i7-4790k to a current system? Ten percent? Twenty....maybe?

I'm already at a dead end at upgrading the system.

Mo memory? I'm already at 32GB for video editing and virtualization.

Storage? I have a 1TB SSD main drive and a 256GB used SSD that is a scratch drive. As for backups and archive files, I have a FreeNAS server with eight 5TB drives in a RAIDZ2 configuration.

Monitor? I have a 27" G-Sync primary monitor running at 2560x1440 resolution, and I have a second 27" 2560x1440 IPS monitor 60Hz monitor "on it's way" for work purposes. No multi-monitor gaming here.

Video card? I have a 980 video card, and I'm looking to replace it with a 1080Ti at the end of the year. Maybe.

The sad truth is that, unless you are a [H]ardGamer or have some [H]ardApplications, the lower-end processors are (G)oodEnough for web browsing and watching videos.

I just upgraded from a Core i7 860 to and Ryzen 1700. That was about 7 years between upgrades. In additional to gaming I do a lot of Handbrake encoding and speeds went up about 4x as expected. But I was also surprised about how much snappier everything felt. Tabbing from a game to windows desktop is now instantaneous. I can even encode and still have no stuttering when playing my games when before it was like playing on a slide show. Yes, the 860 has a slow IPC compared with the Sandy Bridge build and beyond. But it was still a 4 core 8 thread processor with 16 GB of RAM and was also using an SSD drive. So I think the main benefits people will see with these high processor cores is with multitasking. That will be a hard thing for benchmarks to show.

 

[ZLoth]

The improvement you would see would be if you are capping out your 4/8 core/thread cpu. And of course maybe using a new nvme drive. Really does depend on what you are doing but it should come down to price to performance, and right now i dont think Intel has an answer for ryzen.

I have seen actual performance improvement between the two processors. While my motherboard is capable of running a NVMe drive, real-world price performance ratio kinda rules it out for now. (For the record, my FreeNAS box has a 128GB NVMe drive as a boot drive. It was the smallest I could get, and the main reason was not to speed up boot times, but so that I can use all eight SATA ports for the hard drives).

That's nearly 8 years difference between the launches of those CPUs, wait till Q1'22 to make the 4790k to *latest CPU here* speed argument. I completely agree the market stagnated over a few gens, but it seems to slowly be getting better with AMD returning to relevancy in the high end. I made nearly the same leap as you, going from the E6600 to a 3770, then a 4690k because I missed the OCing Will likely stay on this CPU for 3 or 4 more years, unless something comes along and makes meaningful improvements (from a gaming perspective, nothing more).

I typically build my systems so that they last quite a while before they "wear out" or become technically obsolete. That usually is around five years, but because of finances, I had to push my older system for another...years... use. That means that I'm looking at around 2019 for replacement then, and partially for the newer technology.

 

[mat9v]

I was kinda hoping that Ryzen would push Intel more but then I read this on Guru3D:
"Now I did some rounds with the mobo partners and simply asked them what clock frequencies they can tweak the 8 and 10 core parts at with a more normal cooling method, like LCS or a really proper heatpipe cooler. The magic number seems to be 4.2 to 4.3 GHz depending on the ASIC quality. "
It seems that I have no reason to upgrade if, even with water, higher clocks are not possible. I was hoping for 4.7 or even up to 5Ghz on 8 cores and good LCS

 

[Nihilus1]

I just upgraded from a 4790K (OC'd to 4.4GHz) to a 1700X (OC'd to 3.8GHz), and you're right, for basic day-to-day functions, they're indistinguishable. But do you really consider video encoding and photo editing [H]ard? Because with those two tasks, the 1700X blows the 4790K out of the water. And the 1700X has nearly doubled my ability to do bona fide [H]ardApplications required by my scientific research.

Doubling core and thread counts has never made all that much difference for basic stuff - hell, going from my Athlon 64 3000+ to Athlon 64 X2 3800+ only helped with audio and photo editing (I wasn't editing videos back then), and my research applications. It didn't help me type papers or make PowerPoints.

Video encoding and photo editing is a tough argument to make for a new CPU. If processor A does the job in 3 minutes and proccessor B does the job in two minutes, do you really save 1 minute of your time if you are browsing facebook or watching a video clip on youtube anyway. Going to 4/8 threads made a significant difference in productivity. Using SSDs made a HUGE difference in time savings (loading microsoft office especially!) Modern iGPUs handle nearly most non-gaming applications quite well, including 4k playback. Going forward, it will be hard to see the same REAL WORLD improvements, even if performance delta is the same. That is a GOOD thing

 

[egads]

Video encoding and photo editing is a tough argument to make for a new CPU. If processor A does the job in 3 minutes and proccessor B does the job in two minutes, do you really save 1 minute of your time if you are browsing facebook or watching a video clip on youtube anyway. Going to 4/8 threads made a significant difference in productivity. Using SSDs made a HUGE difference in time savings (loading microsoft office especially!) Modern iGPUs handle nearly most non-gaming applications quite well, including 4k playback. Going forward, it will be hard to see the same REAL WORLD improvements, even if performance delta is the same. That is a GOOD thing

Video encoding is probably the best place to see real world improvements. My old i7 860 (a 4 core 8 thread CPU) could encode using handbrake h.264 about 20 fps a 1080p video. For a 2 hour movie that takes 2.4 hours. With the Ryzen 1700 it now can encode around 70 fps cutting the time to 41 minutes, a huge improvement. I'm sure with these new Threadripper and Skylake-E's, it'll be even faster.

Now people are starting to encode in H.265 which is even more CPU intensive. The i7 860 encodes at a pathetic 6 fps meaning 8 hours encoding time for a 2 hour movie. The Ryzen at stock speeds achieves 18.5fps cutting it down to 2 1/2 hours. And for those wondering what overclocking does, I overclocked it to 3.85 Ghz and the RAM to 3200 Mhz and boosted the encoding rage to 23.5 fps or 2 hours. That is why overclocking still matters for even these high end CPUs. I salivate to know how the 18/36 thread beasts can tear through these encodes.

I agree though, in most all other instances, these extra cores won't make much of a difference.

 

[WhoBeDaPlaya]

^ Will echo egads ' sentiments.

My main rig is plenty fine for most uses, but pales in comparison to my two old HP Z800 with dual X5690s.
Those things may be old, but 24c/48t of Westmere goodness still rips encodes a new one!

 

[Peter2k]

I was kinda hoping that Ryzen would push Intel more but then I read this on Guru3D:
"Now I did some rounds with the mobo partners and simply asked them what clock frequencies they can tweak the 8 and 10 core parts at with a more normal cooling method, like LCS or a really proper heatpipe cooler. The magic number seems to be 4.2 to 4.3 GHz depending on the ASIC quality. "
It seems that I have no reason to upgrade if, even with water, higher clocks are not possible. I was hoping for 4.7 or even up to 5Ghz on 8 cores and good LCS

me thinks to remember that all Intel parts are not soldered any more
if true
those chips run into heat problems way before stability problems

well I'm still hoping for coffee lake 6 core parts reaching that high
that would be enough for gaming for the next year's

that being said
I already have read that coffee lake uses 1151 pins like my kaby, but shuffles the pins around
so I would have to buy another board again

if that's true I'll just wait for Ryzen refresh and get me some 8 core part that can clock to maybe 4.5 (yay for pre binned chips)

 

[Dullard]

Video encoding is probably the best place to see real world improvements. My old i7 860 (a 4 core 8 thread CPU) could encode using handbrake h.264 about 20 fps a 1080p video. For a 2 hour movie that takes 2.4 hours. With the Ryzen 1700 it now can encode around 70 fps cutting the time to 41 minutes, a huge improvement. I'm sure with these new Threadripper and Skylake-E's, it'll be even faster.

Now people are starting to encode in H.265 which is even more CPU intensive. The i7 860 encodes at a pathetic 6 fps meaning 8 hours encoding time for a 2 hour movie. The Ryzen at stock speeds achieves 18.5fps cutting it down to 2 1/2 hours. And for those wondering what overclocking does, I overclocked it to 3.85 Ghz and the RAM to 3200 Mhz and boosted the encoding rage to 23.5 fps or 2 hours. That is why overclocking still matters for even these high end CPUs. I salivate to know how the 18/36 thread beasts can tear through these encodes.

I agree though, in most all other instances, these extra cores won't make much of a difference.

It cost more than Ryzen, but the ol' 8 core 5960X has been grinding away at Handbrake for over 2 years. It's nice that you were finally able to join the party!