Intel's i7-7700K Kaby Lake Overclocked and Reviewed at Tom's!

[tbake0155]

On 11/30/2016 at 3:14 AM, patrickjp93 said:

Same with Gallium. Both are toxic heavy metals.

Gallium is far less toxic than mercury.  In fact, it's bordering on being inert under atmospheric conditions.

[MageTank]

9 minutes ago, Lays said:

Wanna see something neat? 

 

http://www.overclock.net/t/1578480/i5-2500k-4-5ghz-vs-6700k-4-5ghz-in-games#post_24549500

 

 

HI @MageTank Suhdude

How long have I been preaching on this forum for people to stop looking at average framerates, and pay attention to minimums? I swear, this will never get through to people, lol. This is the same crowd that claims memory bandwidth does nothing for gaming, because every review ever said so, when not a single one of them bothered to pay attention to CPU overhead and minimum framerates.

[Bouzoo]

21 minutes ago, Lays said:

Wanna see something neat? 

 

http://www.overclock.net/t/1578480/i5-2500k-4-5ghz-vs-6700k-4-5ghz-in-games#post_24549500

 

 

HI @MageTank Suhdude

To be fair, for it to be on the spot it should be 6600K vs 2500k or 2600K (2700K) vs 6700K.

[Guest]

On ‎11‎/‎29‎/‎2016 at 10:52 AM, It's me! said:

The guys at Toms got ahold of an i7-7700K and overclocked it, it really had some heat issues, but the clocks seem pretty decent for what might be early silicon, though they aren't sure if its early or not. It seemed to do well enough though in their full suite, but I'm not sure if its enough to run out and buy one~!!

 

 

 

http://www.tomshardware.com/reviews/intel-kaby-lake-core-i7-7700k-overclocking-performance-review,4836.html

http://www.tomshardware.com/reviews/intel-kaby-lake-core-i7-7700k-overclocking-performance-review,4836.html

This better be the case, because I was really disappointed to see THIS on Intel's website today:  https://drive.google.com/file/d/0BwaQzL0vJPfWc0JfR2xPSFJjSHc/view?usp=sharing

[XenosTech]

26 minutes ago, MageTank said:

How long have I been preaching on this forum for people to stop looking at average framerates, and pay attention to minimums? I swear, this will never get through to people, lol. This is the same crowd that claims memory bandwidth does nothing for gaming, because every review ever said so, when not a single one of them bothered to pay attention to CPU overhead and minimum framerates.

Jeez the minimus are damn near unplayable

[zMeul]

they did a retest: http://www.tomshardware.com/news/intel-core-i7-7700k-kaby-lake-overclocking-update,33119.html - diff mobo with newer BIOS

 

[AresKrieger]

59 minutes ago, MageTank said:

How long have I been preaching on this forum for people to stop looking at average framerates, and pay attention to minimums?

Until those same people blow a couple hundred on a lower end product due to averages

 

But seriously more people need to include things like frame time 1% .1% numbers, mins aren't even enough to me but they are better than straight averages, there is a reason I have an i7 for gaming

[PlayStation 2]

Ivy Bridge part 2 has come to haunt us, I see.

[DELTAprime]

With that retest it sounds more acceptable. Since I'm going from an i5 to an i7 I'll get the 7700k but I wouldn't upgrade if I wasn't jumping up to an i7.

[DELTAprime]

WCCF found a Chinese review of the 7700k. Numbers in a locked 4.0ghz 6700k vs 4.0ghz 7700k test show once again no IPC improvement, I'd say the difference is within margin of error. Also power consumption at temp look within margin of error.

 

http://wccftech.com/intel-core-i7-7700k-vs-core-i7-6700k-cpu-performance-tests/

[patrickjp93]

On 12/4/2016 at 4:00 PM, DELTAprime said:

WCCF found a Chinese review of the 7700k. Numbers in a locked 4.0ghz 6700k vs 4.0ghz 7700k test show once again no IPC improvement, I'd say the difference is within margin of error. Also power consumption at temp look within margin of error.

 

http://wccftech.com/intel-core-i7-7700k-vs-core-i7-6700k-cpu-performance-tests/

Power consumption at one point on the curve does not tell the full story. Skylake takes much more power to hover at 4.5GHz than Kaby Lake does.

[MageTank]

30 minutes ago, patrickjp93 said:

Power consumption at one point on the curve does not tell the full story. Skylake takes much more power to hover at 4.5GHz than Kaby Lake does.

At least you stopped contesting the "zero IPC gain" aspect. That's an improvement. 

[patrickjp93]

1 hour ago, MageTank said:

At least you stopped contesting the "zero IPC gain" aspect. That's an improvement. 

Nope, because there is some according to Agner Fog. The full report will be out in a few weeks, but there are improvements in relative addressing and AVX/2 code paths.

[MageTank]

33 minutes ago, patrickjp93 said:

Nope, because there is some according to Agner Fog. The full report will be out in a few weeks, but there are improvements in relative addressing and AVX/2 code paths.

The Tuesday That Never Came: Volume 2

[porina]

4 hours ago, patrickjp93 said:

Nope, because there is some according to Agner Fog. The full report will be out in a few weeks, but there are improvements in relative addressing and AVX/2 code paths.

Got a link? The update for Kaby Lake isn't available yet, but I take the above statement that some comment was made somewhere on this.

[patrickjp93]

21 minutes ago, porina said:

Got a link? The update for Kaby Lake isn't available yet, but I take the above statement that some comment was made somewhere on this.

We met at a professional developers' conference in Sydney last week hosted by IBM. Preliminary work with the 7700K shows some improvements in IPC, just not in places which affect most people.

[Lays]

3 minutes ago, patrickjp93 said:

We met at a professional developers' conference in Sydney last week hosted by IBM. Preliminary work with the 7700K shows some improvements in IPC, just not in places which affect most people.

in other words:

 no ipc improvement

 

[patrickjp93]

1 minute ago, Lays said:

in other words:

 no ipc improvement

 

No, just IPC where it matters for professional and enterprise software. The crap you're all using isn't even up to date with AVX 1, and that's 6 years old now.

[Lays]

1 minute ago, patrickjp93 said:

No, just IPC where it matters for professional and enterprise software. The crap you're all using isn't even up to date with AVX 1, and that's 6 years old now.

time to ask the real vegetable questions

 

will it increase XTU benchmark score

 

[patrickjp93]

2 minutes ago, Lays said:

time to ask the real vegetable questions

 

will it increase XTU benchmark score

 

Maybe. It has increased Prime95 and AIDA64.

[porina]

4 hours ago, patrickjp93 said:

Maybe. It has increased Prime95 and AIDA64.

You may recall from other threads, Prime95 like loads are what I run 24/7 on my systems, so if this is confirmed it could be big news to me and others like me. Given the mobile versions have been out for a while, I would have thought someone would have noticed by now.

[MageTank]

6 hours ago, patrickjp93 said:

We met at a professional developers' conference in Sydney last week hosted by IBM. Preliminary work with the 7700K shows some improvements in IPC, just not in places which affect most people.

Another secret conference that the general public cannot see, nor can you prove the existence of it. Your tactics have grown stale. Next you will tell me it's obvious, proceed to give me an obscure formula that somehow proves your point (even though it has an infinite number of variables, and makes zero sense) then proclaim that I will see it in due time (which inevitably ends with me being right the entire time, and you avoiding me for months until the next new release comes, and you try again). I've googled "IBM Sydney Conference" and nothing is scheduled for the months of November/December in 2016. Not a single event. Is this super secret event something they wouldn't include alongside the many other conference events on their website?

 

When Kaby finally launches, and it's proven to yield absolutely zero IPC gains (including AVX2, which you magically think will improve), be ready to admit you were wrong. Or do as you normally do, and pretend you "won" every argument ever presented against you. 

[juretrn]

Ahem.

I am not sure if I'm disappointed or what.

 

[porina]

http://www.expreview.com/51056-4.html

 

Another site doing testing on 7700k, and fixing it and 6700k both to 4 GHz, they found... no significant difference. Surprise! There is a fraction of a % here and there but probably down to test variation.

[jagdtigger]

Judging from the single core cinebench test it seems intel still sitting on their money and doing nothing... (it gets only 30 points more than my i5)