Intel Skylake / Kaby Lake

[Khato]

They do:
http://blog.kihltech.com/2014/12/vt...tion_asrock_z97_extreme6_intel_core_i7_4790k/

Looks like you caught me before my quick clarification Since you're quite correct in that a number of motherboard manufacturers have enabled VT-d on the Z series chipsets, but that doesn't change the fact that it's not officially supported by Intel. It's a similar situation to overclocking support on the non-Z series chipsets - not officially supported by Intel, but some motherboard manufacturers have still enabled it.

Edit: Just to be clear, the above is my way of saying that yes, Z170 will likely continue the tradition of not officially supporting VT-d... but that doesn't mean that some motherboards won't work around such just like they do with previous Z series chipsets.

 

[Dave2150]

I'm also waiting until next year after AMD has thrown all of their cards on the table before making a decision on which DDR4 platform to go with X99 Haswell-E or Broadwell-E or Skylake-E / Z170 Skylake or Kaby Lake / Zen AM4.

I suspect prices may not be much lower but there may be some faster memory than the typical 2666 chips on the market now.

In 10 years time, there'll be faster memory available yet again. Maybe wait for that?

 

[PPB]

BCLK OC + such low voltage idle states = recipe for disaster. Manual pstate multiplier control would be ideal for those situations. Disabling offset voltage and going fixed would be a solution but will also defeat the purpose behind all this "5% anual perf gains, but but look at those power savings!" gimmickery.

 

[tenks]

I'm also waiting until next year after AMD has thrown all of their cards on the table before making a decision on which DDR4 platform to go with X99 Haswell-E or Broadwell-E or Skylake-E / Z170 Skylake or Kaby Lake / Zen AM4.

I suspect prices may not be much lower but there may be some faster memory than the typical 2666 chips on the market now.

There already is higher ddr4, up to 3400. Also Skylake-E won't be running on x99 and its at least another 1 1/2 years away. You're always going to be waiting.

 

[Sweepr]

Intel Skylake Non-K Processors Will Feature Overclocking Support With Adjustable BCLK, Memory and iGPU OC

I'm not sure if this is true because it's an ES model in the screenshot (Stepping 1/Q0 Stepping). Even the CPUz 6400T readout could be wrong. Or the BCLK OC restriction wasn't in place in early ES models.

I hope it is. Even if it's not a true return to the old days of great budget Intel CPU OCing every little bit helps. Overclocking the GT2 iGPU should be fun too, especially on cheaper Core i3s.

 

[ShintaiDK]

Looks like you caught me before my quick clarification Since you're quite correct in that a number of motherboard manufacturers have enabled VT-d on the Z series chipsets, but that doesn't change the fact that it's not officially supported by Intel. It's a similar situation to overclocking support on the non-Z series chipsets - not officially supported by Intel, but some motherboard manufacturers have still enabled it.

Edit: Just to be clear, the above is my way of saying that yes, Z170 will likely continue the tradition of not officially supporting VT-d... but that doesn't mean that some motherboards won't work around such just like they do with previous Z series chipsets.

I think like previous, the biggest issue will not be CPU or chipset support. But rather finding a mobo with BIOS support.

People with proper CPU, proper chipset(Official terms) would still find mobos not supporting it due to lacking BIOS.

But some Z170 mobos as you say will support VT-D (un)officially. Asrock will for sure be one of them

 

[DrMrLordX]

Is it possible that Skylake non-K processors will allow bclk strap OC on boards that support it? The benchlife screenie showed a bclk of 133 mhz, which seems to support that hypothesis. Had we seen an atypical bclk speed (such as 122 mhz, or 145 mhz, or whatever) then we could conclude that hand-tuned bclk is possible, at least for an ES.

 

[rtsurfer]

I'm also waiting until next year after AMD has thrown all of their cards on the table before making a decision on which DDR4 platform to go with X99 Haswell-E or Broadwell-E or Skylake-E / Z170 Skylake or Kaby Lake / Zen AM4.

I suspect prices may not be much lower but there may be some faster memory than the typical 2666 chips on the market now.

There will be 3800Mhz+ memory kits available at Skylake launch, they will cost a pretty penny though.

 

[escrow4]

If its not meant to be overclocked and isn't K why are you fiddling with it? Its like you are asking for issues.

 

[nenforcer]

In 10 years time, there'll be faster memory available yet again. Maybe wait for that?

Who are you - the marketing manager for Kingston? Are poor quarterly sales results forcing you to peddle new chips to posters on tech website forums?

Your missing the point - for people with a fairly recent computer with high speed 1866 DDR3 memory like myself the performance advantage of DDR4 isn't that great currently at DDR4 2666 speeds. It's just better to wait until AMD's chips are on the table before deciding which system to build Zen or whatever Intel has out next year when it releases.

 

[dark zero]

Nehalem all over again?
If that so... Kabylake would be the new Sandy Bridge!

BTW, that would kill the whole Pentium K prices.

 

[Sweepr]

http://edm.msy.com.au/24072015/intel/online/html.html

If there isn't enough evidence about next week's launch.

 

[Zodiark1593]

When the processors launch on the 5th, will they be available to order right away, or does it take a few days?

 

[iSkylaker]

When the processors launch on the 5th, will they be available to order right away, or does it take a few days?

I remember Haswell launch they were available to buy on Newegg the next or same day after launch. Maybe SKL gets availability right away unlike Broadwell.

 

[dark zero]

Mayeb Skylake was finished and are already on the stores like the few surviving Carrizo laptops and the H Broadwell ones being ready to being sold... just like the Apple iPhone fashion.

 

[mikk]

Interesting, the Carrizo A10 8700P notebook tested by NBC managed only 1510 points @ 3DMark 11 Performance. Even if it was set at 15W TDP (a bit strange for a 17'' notebook where thermal headroom and battery sizes are not exactly an issue) that's still considerably slower AMD's marketing slides:

AMD's Marketing

Actual OEM Notebook
www.notebookcheck.net/AMD-Radeon-R6-Carrizo-Benchmarks.144290.0.html

I wonder what went wrong going from the AMD seal of approval reference platform to the actual OEM notebook? This has left me a bit worried, after the disappointing CPU performance I was expecting at least a significant graphics performance advantage. Skylake-U GT2 could score north of 1650 points (dual-channel) if the estimates prove to be true and there's a Skylake-U GT3e SKU coming too.

I have asked Notebookcheck and they confirmed me this device is running 2x4 GB Dualchannel. Their article is not yet finished.

Not looking too good, as expected AMDs 3dmark marketing slides are misleading. Compared with HD 5500: http://www.notebookcheck.com/Intel-HD-Graphics-5500.125585.0.html
(make sure you are comparing the better dualchannel results)

There is only a big difference in 3dmark. Basically all other benchmarks from Cinebench OpenGL, gfxbench and most gaming benchmarks is a close match. In some it is even losing to HD5500. And this is only Broadwell.

 

[Sweepr]

I have asked Notebookcheck and they confirmed me this device is running 2x4 GB Dualchannel. Their article is not yet finished.

Not looking too good, as expected AMDs 3dmark marketing slides are misleading. Compared with HD 5500: http://www.notebookcheck.com/Intel-HD-Graphics-5500.125585.0.html
(make sure you are comparing the better dualchannel results)

There is only a big difference in 3dmark. Basically all other benchmarks from Cinebench OpenGL, gfxbench and most gaming benchmarks is a close match. In some it is even losing to HD5500. And this is only Broadwell.

Interesting, thanks for clarifying mikk.
Given AMD's track record only a fool would believe the lies about better CPU performance than Broadwell-U or Skylake-U but I never expected such a small difference when it comes to graphics. If Skylake's Gen 9 iGPU meets Intel's expectations things could get ugly, not sure if AMD can survive without a new mobile APU till 2017.

 

[dark zero]

If that is true, also any nVIDIA chip up to GTX 960M will be doomed against Intel Solution (their Iris Pro, right now they are dooming up to 950) leading into a massive AMD & nVIDIA lose of terrain in that aspect...
I was optimist before since both could handle the Notebook graphics market up the next generation, where their true replacements will come, but if Intel is improving that great, I am fearing that even NVIDIA with HBM will come too late to the party

 

[mikk]

Interesting, thanks for clarifying mikk.
Given AMD's track record only a fool would believe the lies about better CPU performance than Broadwell-U or Skylake-U but I never expected such a small difference when it comes to graphics. If Skylake's Gen 9 iGPU meets Intel's expectations things could get ugly, not sure if AMD can survive without a new mobile APU till 2017.

In the past Intel did better in 3dmark (especially Gen7 against AMDs VLIW), it's a turnaround with later GCN versions, they tendentially do better in 3dmark11 (that's why AMD likes 3dmark11 slides so much) compared to Intel.

 

[majord]

I have asked Notebookcheck and they confirmed me this device is running 2x4 GB Dualchannel. Their article is not yet finished.

Not looking too good, as expected AMDs 3dmark marketing slides are misleading. Compared with HD 5500: http://www.notebookcheck.com/Intel-HD-Graphics-5500.125585.0.html
(make sure you are comparing the better dualchannel results)

There is only a big difference in 3dmark. Basically all other benchmarks from Cinebench OpenGL, gfxbench and most gaming benchmarks is a close match. In some it is even losing to HD5500. And this is only Broadwell.

Firstly, I wouldn't call that a close match. Against HD6000, yes it trades blows, but then you'd need to find an R7 to compare to instead.

As for the score discrepency with AMD's slides.. well, have you actually looked at the results on NBC for the core U? If so, have you not noticed the delta between the many different implementations? it's as much as 100% on gaming tests! ,far far more than LP single ch vs dual ch alone can account for. It's clearly very difficult to get anywhere close to consistent results across machines with any low power, dynamically clocked CPU/APU. particularly in gaming results due to the CPU+GPU loading

Just look at the Core M debacle (Which wouldn't be a debacle at all if they wern't trying to push it into such a low TDP/SDP which it's really not suitable for)

In a nutshell , don't write off Carrizo based on a single test result.. especially one that's just ~10% off the 'ideal' Reference system.

It's pretty obvious it's running 12-15w TDP, given it aligns closest to the expected results floating around, and given that, even it's CPU performance isn't that far off the mark. Much like how BW GT2 isn't far off the mark in it's graphics performance.

I think BOTH these facts are a surprise, and I tip my hat to both companies really in each respect.

 

[mikk]

Firstly, I wouldn't call that a close match.

There are lots of benchmarks where it is close or losing to HD5500.

As for the score discrepency with AMD's slides.. well, have you actually looked at the results on NBC for the core U? If so, have you not noticed the delta between the many different implementations? it's as much as 100% on gaming tests! ,far far more than LP single ch vs dual ch alone can account for.

For Intel not only SC vs DC but also different SKUs with different Turbo specs. And regarding gaming benchmarks, in some games there is only one dualchannel HD5500 device in comparison which possibly wasn't the best for HD5500 either. And other gaming benchmarks with more results are pretty consistent. Not much difference there between HD5500 dualchannel devices.

Just look at the Core M debacle (Which wouldn't be a debacle at all if they wern't trying to push it into such a low TDP/SDP which it's really not suitable for)

This Carrizo is shrouded in a big 17" notebook with a rather low TDP by the looks of it, I don't think it is comparable. Regardless of this AMD does much better in 3dmark11. There was a reason we asked for tests. This is the first and it wasn't unexpected that it doesn't look as good.

 

[Sweepr]

In a nutshell , don't write off Carrizo based on a single test result.. especially one that's just ~10% off the 'ideal' Reference system.

The score is worse than what AMD promised for the low-end A8 8600 at 15W, it is worrying.

It's pretty obvious it's running 12-15w TDP, given it aligns closest to the expected results floating around, and given that, even it's CPU performance isn't that far off the mark. Much like how BW GT2 isn't far off the mark in it's graphics performance.

Is it? Cause I don't see any tests, you're assuming this because the disappointing scores are closer to what AMD promised at 15W (reference platform) but it could very well be running at 20W or more inside a 17'' notebook. The significant ST performance gap from previous generations is still there and they failed to establish a MT performance advantage, hence why CMT is dead and buried for their future mobile APUs. Looking at Llano vs Sandy Bridge just a few years ago their graphics performance lead is eroding fast too.

 

[Sweepr]

When the processors launch on the 5th, will they be available to order right away, or does it take a few days?

They will, but supply could be short initially.

Intel Skylake CPUs may be in short supply at launch

Intel will soon be launching their Skylake CPUs and their new Z170 Socket motherboards, with a rumored launch date of August 5th at Gamescom.

Today we have reports from the Swedish retailer Wabhallen that Intel's upcoming CPUs will be in short supply, saying that initial deliveries of Intel's i7 6700K and i5 6600K CPUs will not be enough to last for even a week of sales. Below is a quote from Webhallen's marketing director Anton Nilsson.

We have a small number of processors and motherboards coming. They confirmed deliveries will not be enough for a whole week, rather a few days.Unfortunately, it also seems that those of our PC series most interesting motherboard is further delayed about a month,

www.overclock3d.net/articles/cpu_mainboard/intel_skylake_cpus_may_be_in_short_supply_at_launch/1

AIDA64 v5.30 Press Release
Microsoft Windows 10 and Intel Skylake Support

www.aida64.com/news/pr/aida64-v530-press-release

After confirming published August 5, now Japan has confirmed the release date sixth generation Intel Core platform.

Intel announced the sixth generation of the Japanese auto Core, which is the release date Skylake platform. Expected Z170 processor motherboard with Skylake in Japan local time on August 5 21:00 officially on sale.

http://benchlife.info/japan-announce-skylake-platform-sell-timing-in-twitter-07282015

 

[ShintaiDK]

Off topic, but for those wanting Windows 10 now and got it reserved. You can do so:
http://forums.anandtech.com/showpost.php?p=37588924&postcount=3

 

[Zodiark1593]

They will, but supply could be short initially.

Intel Skylake CPUs may be in short supply at launch



www.overclock3d.net/articles/cpu_mainboard/intel_skylake_cpus_may_be_in_short_supply_at_launch/1

AIDA64 v5.30 Press Release
Microsoft Windows 10 and Intel Skylake Support

www.aida64.com/news/pr/aida64-v530-press-release

http://benchlife.info/japan-announce-skylake-platform-sell-timing-in-twitter-07282015

Yikes, so I will want to order both mobo and processor ASAP on the 5th then, perhaps at midnight.