CHADBOGA
Platinum Member
- Mar 31, 2009
- 2,135
- 832
- 136
They could have done that long ago. None of us wanted the IGP anyway.
I do.
That way if a new GPU I want is a few months away, I can make do with the IGP until then.
I do.
That way if a new GPU I want is a few months away, I can make do with the IGP until then.
LTC8K6
Lifer
- Mar 10, 2004
- 28,520
- 1,575
- 126
You must not be one of us, then.
You could make do with your old video card, too.
I keep a couple 1GB GDDR5 GT730 cards around as stand ins / test cards.
The funny thing is the IGP is still there in the 7740K, it's just disabled.
With the 7700K we were paying for an IGP we didn't want, but at least we could use it if we wanted to.
Now if we bought a 7740K, we'd be paying for an IGP that's disabled and unusable.
Arachnotronic
Lifer
- Mar 10, 2006
- 11,715
- 2,012
- 126
True, but in exchange we seem to get more frequency headroom thanks to the superior packaging.
TahoeDust
Senior member
- Nov 29, 2011
- 557
- 404
- 136
If you pre-ordered your chip from B&H when their prices were inflated yesterday morning, their prices have been changed to fall in-line with everyone else. Shoot them an email or give them a call and they will adjust your price and/or refund you the difference.
CHADBOGA
Platinum Member
- Mar 31, 2009
- 2,135
- 832
- 136
I gave it to a friend and am now using the IGP, so when I buy my next system, options are good.You must not be one of us, then.
You could make do with your old video card, too.
TahoeDust
Senior member
- Nov 29, 2011
- 557
- 404
- 136
An interesting bit from TechReport's review:
Maybe I'm just cynical, but I read this as meaning that the transistor density would reveal that Intel's 14nm++ process really isn't that much more dense (if at all) than foundry 14nm. Consider this recent article, in which Intel argued that their 14nm process has almost as much transistor density as the upcoming foundry 10nm processes. The article contains extensive claims and discussions about density as measured in MTr/mm^2, but now that Intel has the opportunity to give the figures for an actual shipping product, they don't. I wonder if this is because the hypothetical transistor density figures apply to something simple like DRAM, and that CPUs engineered for high clock speeds need to space things out a bit more. I suspect the real transistor density for Skylake-X is roughly on par with that of Ryzen.
As for characteristics of the Skylake-X silicon itself, Intel honchos clammed up when we asked about die size and transistor count. The company believes that disclosing this information will lead to unfounded conclusions from its competitors about the quality of their chip designs and process technologies compared to Intel's. Only the paranoid survive, we suppose.
Maybe I'm just cynical, but I read this as meaning that the transistor density would reveal that Intel's 14nm++ process really isn't that much more dense (if at all) than foundry 14nm. Consider this recent article, in which Intel argued that their 14nm process has almost as much transistor density as the upcoming foundry 10nm processes. The article contains extensive claims and discussions about density as measured in MTr/mm^2, but now that Intel has the opportunity to give the figures for an actual shipping product, they don't. I wonder if this is because the hypothetical transistor density figures apply to something simple like DRAM, and that CPUs engineered for high clock speeds need to space things out a bit more. I suspect the real transistor density for Skylake-X is roughly on par with that of Ryzen.
If you paid attention in the past you would notice that Intel either counts it's transistors differently, lies about density or does something very strange with design of their chips. Because their transistor counts are consistently and significantly lower than what AMD claims for it's own CPUs. Like, if we are to believe AMD's and Intel's claims alike, you would learn that Ryzen has about as many transistors as 24 core Broadwell or something silly like that. So, from my perspective Intel's excuse not to disclose that is valid, because in the end it will just lead to certain folk claiming BS about Intel's process.
As for die size we already know it to be about ~1.5 times the KBL-S (built on same process) one for LCC die. Make of that what you will. And yeah, density wise they are not too different, but density here is defined by design, not by process characteristics and the only real density comparison you can make here is to take look at each die and figure out how much space cache occupies.
Last edited:
ZGR
Platinum Member
- Oct 26, 2012
- 2,052
- 656
- 136
You must not be one of us, then.
You could make do with your old video card, too.
I keep a couple 1GB GDDR5 GT730 cards around as stand ins / test cards.
The funny thing is the IGP is still there in the 7740K, it's just disabled.
With the 7700K we were paying for an IGP we didn't want, but at least we could use it if we wanted to.
Now if we bought a 7740K, we'd be paying for an IGP that's disabled and unusable.
Intel will hopefully enable them somehow. It is amazing how useful an iGPU is.
I would have agreed would you if it were not for the i7-5775c's insane iGPU. Now I want all my next CPU's to have a backup iGPU to have this much grunt.
It is great being able to play old RTS games at max settings at 2160p on an APU so to speak when you don't have a GPU.
My old i5-4690k certainly couldn't do that.
Do I need an iGPU? No, but it certainly is nice to have a fast one.
I sure hope Intel keeps at it with the fast iGPU's with their K processors, if only to add that sweet L4 cache.
Yet another Coffee Lake-H submission @ GFXBench. Makes me wonder if some mobile parts could be released earlier than expected as well.
From SiSoftware a few days ago: Genuine Intel(R) CPU 0000 @ 2.60GHz (6C 12T 2.59GHz, 6x 256kB L2, 9MB L3) (Intel CoffeeLake H DDR4 RVP)
The fact that they already have an ES with 6C/12T at 2.6 GHz base @ 45W TDP (standard for H SKUs) is good news. Core i7-7700HQ, probably the most popular Kaby Lake-H CPU, packs 4C/8T at 2.8 GHz @ same TDP.
From SiSoftware a few days ago: Genuine Intel(R) CPU 0000 @ 2.60GHz (6C 12T 2.59GHz, 6x 256kB L2, 9MB L3) (Intel CoffeeLake H DDR4 RVP)
The fact that they already have an ES with 6C/12T at 2.6 GHz base @ 45W TDP (standard for H SKUs) is good news. Core i7-7700HQ, probably the most popular Kaby Lake-H CPU, packs 4C/8T at 2.8 GHz @ same TDP.
LTC8K6
Lifer
- Mar 10, 2004
- 28,520
- 1,575
- 126
You really shouldn't use the 5775C as an example, since it's pretty much unique with the 128mb of L4 cache and 48EUs. No other desktop Intel IGP had that, and so no other desktop Intel IGP compares to it. Iris Pro 6200 is much faster overall than HD630, for example. 48EUs and 128mb of EDRAM vs 24EUs and no EDRAM.
Even so, the IP6200 IGP is still handily defeated by most low end cards, as it is always hampered by sharing DDR3 ram.
Arachnotronic
Lifer
- Mar 10, 2006
- 11,715
- 2,012
- 126
Interesting that this is a 6C/12T SKU with 9MB L3 cache. I wonder if Intel will segment the mobile parts by L3 cache size within the i7 line...
I hope this means we have the Core i5 as only quad cores and the Core i3 as only dual cores now - I have had to explain to many people that a mobile Core i7 is not always a quad core.
Like mobile i7s with 6MB L3? Looks like business as usual to me.
ZGR
Platinum Member
- Oct 26, 2012
- 2,052
- 656
- 136
The 5775c's successor was Skull Canyon; remember the NUC? it isn't fair to say it shouldn't count. Intel is absolutely expanding their Iris lineup, it just is moving away from desktop it seems.
They really need to release its successor! The 5775c was Intel's costly answer to AMD's APU's. There's a reason I'm still using my 5775c. But any recent 6700HQ/7700HQ have 128 MB L4 and the Iris GPU. And so do the Xeons with Iris. So really, the 5775c is just a small part of the ever expanding L4 lineup.
Intel has the CFL mobile 6c12t Iris L4 it appears so lets hope they can bring it to Desktop. I will happily buy that instead of going SKL-X; if it can overclock well that is.
And Intel has segmented mobile i7's by L3 cache for many years. It was 6 MB for the low end 4c/8t i7's and 8 MB for higher clocked 4c/8t models. The sizes changed slightly with Skylake.
Last edited:
I really hope that desktop i7 6/12 8700K (or whatever name will have) will be at least on a clock level with Skylake (4.0-4.2Ghz) with easy overclock up to 4.5-4.6Ghz. Hopefully I don't have a very high expectations.
LTC8K6
Lifer
- Mar 10, 2004
- 28,520
- 1,575
- 126
6700HQ and 7700HQ have standard HD630 graphics. 24EUs and no L4 cache.
IP 5200 40EU 128mb mobile/server
IP 6200/IP P6300 48EU 128mb 2 desktop chips - rest mobile/server
IRIS 540/640 48EU 64mb mobile
IRIS 550/650 48EU 64mb mobile
IP P555 48EU 128mb server
IP P580 72EU 128mb server
IP 580 72EU 128mb mobile
IP = Iris Pro
I think that covers all the IGPs with EDRAM L4.
I believe there are only two desktop CPUs in there, 5775C and 5675C with IP6200.
Only the 580 variants have 72 EUs and 128mb of EDRAM, and are the fastest IGPs of the bunch.
https://en.wikipedia.org/wiki/Intel_HD_and_Iris_Graphics
TheF34RChannel
Senior member
- May 18, 2017
- 786
- 309
- 136
I was just about to opt something similar when I read Sweepr's post. What I could really use is some -S entries/leaks...
Pretty sure you'll see that wish granted, the SiSoftSandra entry for the ES -S part showed a 4.2GHz clock and that's not necessarily finalized clocks either. I wouldn't worry about it honestly.
- Sep 28, 2005
- 20,841
- 3,189
- 126
Why? The only purpose for IGP on skylake-X would be a power save feature where it puts the gpu's in low power mode... and runs it off IGP.. (oh wait nvidia already does that...)
So no... IGP is pointless on skylake-x.
Unless your a virgin purchaser, meaning u have nothing to recycle, you should have a previous gen GPU to sit on.
So any guess when mature boards and 7920 cpu will be out?
I have been a victim of beta boards, pre release boards, and launch day hashs, and well, i can say intel sucks at launch for anything besides full bloated enterprise.
So im waiting for the mature stuff, the boards laced with options, and bios tweaks preinstalled so i dont need to worry about bricking the thing from a power outage or some unforeseen mahem of god.
ZGR
Platinum Member
- Oct 26, 2012
- 2,052
- 656
- 136
My bad, I mixed up R and HQ. There are several Desktop CPU's with L4, albeit at limited release; but they do exist. I believe you are thinking of unlocked CPU's with L4?
List of "Desktop" CPU's with L4 Cache:
Haswell
- i7-4770R
- i5-4670R
- Xeon 1284LV3
Broadwell
- i7-5775C
- i5-5675C
- i5-5575C
- Xeon 1284Lv4
- Xeon 1278Lv4
Skylake
- i7-6785R
- i5-6685R
- i5-6585R
Skylake differs in that there are no desktop Xeons with Iris Pro? Frankly, I don't care too much about the iGPU past the point that it can drive 4k60 content.
I care more about using the L4 cache as a victim cache to the L3 cache. Skylake introduced more ways of utilizing the L4 cache as well. It is prime time to have another "halo" Iris product for mainstream.
Edit: I use my iGPU mainly for OS X. Without my iGPU, upgrading to my 1070 would have been a nightmare as Pascal support was non existant for a year. I hate adding additional drivers to my OS X build for fear of kernel panics (old fear) so using built in HD driver kexts was such an easy fix.
Ajay
Lifer
- Jan 8, 2001
- 15,406
- 7,833
- 136
What has it usually been before v.2 boards appear - six months or so? Or, will mobo makers wait a year or so till Cascade Lake X is out?
At least that i9 7920 will really put some hurt on a custom cooling loop.
TRENDING THREADS
-
Discussion Intel current and future Lakes & Rapids thread
- Started by TheF34RChannel
- Replies: 23K
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 9K
-
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 7K
-
Facebook
Twitter