Move over 7700, i3 8350 almost matches in performance

[Taja]

2 hours ago, mr moose said:

obviously?  So you have in-depth experience and understanding of the R+D division of Intel?  

Oh well, surely with this 'argument' you destroy any kind of conversation. but yeah sure, every one believes what they want at the end of the day

[mr moose]

7 minutes ago, Taja said:

Oh well, surely with this 'argument' you destroy any kind of conversation. but yeah sure, every one believes what they want at the end of the day

well what you were saying goes against observation.  So I assumed you had inside knowledge.

 

EDIT: But I will elaborate,  When a company has no competition (like Intel and nvidia over the past few years), it gives them time to polish their products a bit more before release.  If Intel truly had a stock pile of tech already developed but just not releasing it due to lack of competition, then when they brought forward the x299 release, we would have seen a better finished product.  Instead we didn't, and the reason we didn't was because it was only just being finished as they were bringing it forward.   Ergo they are not holding back development.    The other major reason why they don't do what you think they do, is because these companies can't rely on new sales,  all new markets hit a peak and then stagnate or even drop off.  Companies like Intel are reliant on return custom for ongoing revenue, this will only happen if they give their consumers a reason to buy a new product, ergo they have to keep developing and releasing.  This concept that companies delay development, don't try or even develop but don't release is just a fallacy of the internet. 

 

This is not to even mention the the fact that complacency does nothing more than give your competition a chance to make a better product.

 

 

[Cookybiscuit]

I'm still so mad they didn't even announce a release date on their stupid conference on the 21st, Brian must not understand people are waiting, how rude.

[Chaos_Sorcerer]

If this is true, it either costs an absurd price or has like 2 PCIe lanes. 

[Lays]

18 hours ago, MyName13 said:

But CL i3 won't have HT.

Weird, didn't realize you knew every single CFL sku, you have no clue what they're going to offer lol

[Lays]

2 hours ago, mr moose said:

obviously?  So you have in-depth experience and understanding of the R+D division of Intel?  

I wouldn't listen to him

[MageTank]

18 minutes ago, Lays said:

Weird, didn't realize you knew every single CFL sku, you have no clue what they're going to offer lol

You don't know what you are talking about. 

[Lays]

Just now, MageTank said:

You don't know what you are talking about. 

or do i

 

 

:]

[MageTank]

Just now, Lays said:

or do i

 

 

:]

My dad works at Intel.

[Lays]

Just now, MageTank said:

My dad works at Intel.

[ARikozuM]

Just now, MageTank said:

My dad works at Intel.

My mother works for AMD. /s

[MageTank]

1 minute ago, Lays said:

[Lays]

Just now, MageTank said:

[MageTank]

Just now, Lays said:

[Lays]

1 minute ago, MageTank said:

[MageTank]

Just now, Lays said:

[RGProductions]

but it will cost $200 

/s

[mr.squishy]

20 hours ago, FloRolf said:

If I bought a new i7 within the last 2 years I'd be mad lol. 

Good thing my 2600k still lives. 

I got a 6700k for $200 and I'm still gonna be mad if this is true.

[porina]

8 hours ago, MageTank said:

The CPU-Z result can be ignored. You cannot know the speed of the CPU, because it only reports the DMI string. I can show you my CPU-Z screenshot at 4ghz, restart, and dial in 5ghz (it's a K SKU, meaning it's multiplier is easily adjustable even in XTU) and score higher. 

I was making the assumption (not proven) that the same source did both the CPU-Z and Aida64 runs, at the same settings.

8 hours ago, MageTank said:

The AIDA 64 benches mean absolutely nothing to me. Photoworxx is NOT a CPU test. It's for ram. I can run a dual core CPU and put i7's to shame. In fact, I've done so in the past: 

Aida64 "CPU" benches don't mean anything to me either, for different reasons. I just never used them for anything. With limitations and unknowns, I still tried to replicate the conditions of the reported results.

8 hours ago, MageTank said:

I just ran the numbers on my 7700k with HyperThreading turned off. At a x40 multiplier and 103 BCLK, I am hitting 4120mhz. I am scoring around 480 points single, 1860 multi in CPU-Z. Cache is still 8mb. I am also using heavily tweaked ram at 2472mhz (C10-10-10-28-1 with 190 tRFC and 65535 tREFI, tRDWR at 10 as well). You cannot tell me their ram was as tweaked as mine during their test, and that they still outscored be my nearly 7-8% on IPC alone. The numbers just don't add up.

So similar numbers to mine, absent info on if the bench is influenced by cache/ram speeds/sizes.

 

Possible scenarios are the leaded results are: from a higher clock than we think, have a better IPC than we think - from any of CPU, cache, ram improvements. Gonna be interesting finding out. I'm almost certainly going to get a 6 core to test when available.

[Sauron]

9 hours ago, ARikozuM said:

Even games and editing suites, i.e. CAD or Adobe, still rely heavily on single-threaded performance.

Again, why would you buy a 400$ 6 core for that?

[The Benjamins]

5 minutes ago, Sauron said:

Again, why would you buy a 400$ 6 core for that?

just for some info, I am a drafter and I use CAD programs at work but to do my job I usually have 4-10 other programs open so each program might not take full advantage of the CPU but the cluster of them can.

 

But also I never run into CPU bottle necks even on old i5 CPUs.

[That_PC_Kid]

The CL i3's look very promising. I plan on making a build around the i3-8350K but I might move up to an CL i5 if it is not that much of a price jump.

[Taja]

12 hours ago, mr moose said:

  Ergo they are not holding back development.   Companies like Intel are reliant on return custom for ongoing revenue, this will only happen if they give their consumers a reason to buy a new product, ergo they have to keep developing and releasing.  This concept that companies delay development, don't try or even develop but don't release is just a fallacy of the internet. 

 

This is not to even mention the the fact that complacency does nothing more than give your competition a chance to make a better product.

 

 

Maybe not holding back developmet, but holding back the PRODUCT. Selling 8 core for 1000$, and than for 600$, and now for less than 400$... They had the product, they just overpriced it TO SHIT, and created a product line that only makes sense because they had no competition. now that they do, they can just change their strategies, they dont need to make heavy investment.

 

and about thew development part, they were focusing on other areas, because they launched "compelling" products every year (if you call 10% compelling), and in the market they were already kings, they didnt make a lot of effort.

 

And to finish, their complacency DID gave AMD space, but if intel does what they seen to be doing now (with coffee lake, not the shitlakes-x mess), they will be just fine.

[MageTank]

10 hours ago, porina said:

I was making the assumption (not proven) that the same source did both the CPU-Z and Aida64 runs, at the same settings.

Aida64 "CPU" benches don't mean anything to me either, for different reasons. I just never used them for anything. With limitations and unknowns, I still tried to replicate the conditions of the reported results.

So similar numbers to mine, absent info on if the bench is influenced by cache/ram speeds/sizes.

 

Possible scenarios are the leaded results are: from a higher clock than we think, have a better IPC than we think - from any of CPU, cache, ram improvements. Gonna be interesting finding out. I'm almost certainly going to get a 6 core to test when available.

I can rule out ram improvements. There is no way their 2472 C11-13-13-28-2 beat my 2472 C10-10-10-28-1 with extremely tight tRFC and tertiary timings. Cache, due to it's very nature, doesn't often exceed core clocks, so if they were indeed running 4120mhz, they were using the same cache clocks that I had. 

 

Even looking at the photoworxx result, their memory bandwidth seems off. It should be about 22k, not 25k. I cannot tell if they are blatantly lying about the numbers, whether or not we are dealing with a photoshop, or if they somehow managed to improve upon this generation without changing the architecture at all. Another reason I am highly skeptical, is in those Aida 64  screenshots, the CPU is listed as "Devil's Canyon". Could be a fluke, could be something Intel added during their testing phase of the CPU, or perhaps it's a software glitch with Aida, I just do not see the point of adding that to the CPU's DMI string at all. 

[iamdarkyoshi]

On 24/08/2017 at 3:36 AM, FloRolf said:

If I bought a new i7 within the last 2 years I'd be mad lol. 

Good thing my 2600k still lives. 

Me with my 6700K

 

Whatever, its delidded which gets me a hell of a lot more performance