• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Ryzen "2" ?

I definitely see the point - it is an interesting comparison to understand the differences in the different architectures.

It remains to be seen if it is the silicon fabrication or something about the chip architecture that is limiting AMD to max of 4.3ish not the 5+ Ghz that intel can achieve - but it is still information that is interesting.

I think we simply need to have a slight change of definition to IPSC (instructions per single core) and IPCC (Instructions per combined Cores). I think that comparisons of both are perfectly valid. Today we have processes that are still thread limited - but with both Intel and AMD going to higher thread counts that gap should continue to diminish. So the technology to measure how well a chip can use it's full capacity at a given speed is relevant today and will be more relevant moving forward.

Measuring IPSC, IPCC, total Mhz achievable and scaling to Memory speed lets us know where each player is doing best and what a consumer ought to buy for their specific needs.

LOLN1: Could you confirm what memory you are running and what the maximum stable memory speed & timings is that you are able to achieve? It looks slower than I was expecting.
 
Last edited:
DragonQ said:
This really makes no sense. How is comparing "SMT A" to "the lack of SMT A" in any way similar to comparing "SMT A" to "SMT B"?

Maybe the term "IPC" wasn't used but everyone made comparisons between Core i5s and Core i7s, with SMT obviously being the differentiator. IPC doesn't inherently mean "per core", although that's also a legitimate comparison. IPC is just that, instructions per clock, and you can do more if the application uses SMT effectively.

When most people use the term IPC they do really mean "IPC per core" though. It's an odd one because a "core" constitutes lots of parts, and you wouldn't say "it's not a valid IPC comparison because one chip has more cache" or "it's not a valid IPC comparison because one chip has AVX512 instruction set and the other doesn't". So I feel SMT is a perfectly valid variable to include, it just obviously makes no difference to single threaded workloads.
Click to expand...

yeah IPC term is overused, we should be using PPC, performance per core.

e.g. if ryzen 2 was 6ghz but with say 95% of intels IPC, it would have better PPC due to the higher clock speed and as such be the winner on single threaded software.

Also as you said there is other factors such as cache, imc performance etc.
 
Thanks to the guys who posted some single threaded Ryzen scores for me. Although Cinebench isn't the only way to measure performance it does give a good impression of performance and I know roughly where things stand now. Sadly single threaded even on the 2700 is still behind my 4790k (single core score 190) and while they would destroy my cpu in multi threaded (score 946) the performance I need is mostly single threaded so still no upgrade for me. I don't think I have ever had the same platform for so long.
 
pastymuncher said:
Thanks to the guys who posted some single threaded Ryzen scores for me. Although Cinebench isn't the only way to measure performance it does give a good impression of performance and I know roughly where things stand now. Sadly single threaded even on the 2700 is still behind my 4790k (single core score 190) and while they would destroy my cpu in multi threaded (score 946) the performance I need is mostly single threaded so still no upgrade for me. I don't think I have ever had the same platform for so long.
Click to expand...

Like all benches, it just shows how well it is at doing that specific job and a rough guide to general performance.

As I said over a year ago now (Jesus, that went quick), gaming is noticeably smoother, multicore performance is good, it's cool and quite and is priced very well. It depends what's more important to you.

I have no intention of replacing the 1800X in my main PC as I got a damn good cool chip and wouldn't be worth the cost of upgrading this year. Next year will be an overhaul providing AMD deliver.
My other two (1600X and 1800X) I'll upgrade for work so I can play with them.

edit: Though having said that, I would be tempted if they released a 2800X using golden sample chips. So who knows, i'm fickle.
 
Last edited:
chrcoluk said:
yeah IPC term is overused, we should be using PPC, performance per core.

e.g. if ryzen 2 was 6ghz but with say 95% of intels IPC, it would have better PPC due to the higher clock speed and as such be the winner on single threaded software.

Also as you said there is other factors such as cache, imc performance etc.
Click to expand...

The problem with that is "Percore performance" is not that black and white, you can't say "this is the performance of one core on Intel vs Ryzen so just multiply the number of cores" that's completely flawed because it ignores the fact that Ryzen has better multicore performance, so the multicore performance is faster relative to Intel than one core performance, if you ignore that you're understating the performance of Ryzen relative to Intel.
I'm not talking about "Ryzen has more cores" with the same number of cores/threads Ryzen is faster relative to Intel in multicore than it is just restricted to one single core, its why some people like the idea of reinventing measuring performance just using a singular core.
 
Last edited:
Out of interest how well would a clocked 12/1400 compare to a stock 3570k? Wondering as this Mobo is slowly turning up its toes... And will there be a quad available in the Ryzen 2 cpus in a couple of weeks?
 
Zeed said:
Anyhow I think Zen+ looks Good for what it IS an Evolution of Zen. I think AMD is making mistake calling it Ryzen 2 should stick to Ryzen Plus or Ryzen Refresh. So people would not get tooo hyped.

but its AMD masters of HYPEEEE TRAINNN that usually derails :p
Click to expand...

Why have you become hyped? Everyone seems to be patiently waiting for the release.
 
It seems fairly balanced going off some of these early results depending what you want out of your PC. If this refresh was more like 4.3-4.6 I'm sure a lot of people would sacrifice clock speed for two extra cores providing fast memory is also easy to run. Still seems Intel is a safer bet unless you do heavy workloads, are on a budget or heavily multitask/stream. Then again, also wish Intel wouldn't use the toothpaste they have done on 8700k.
 
Drollic said:
It seems fairly balanced going off some of these early results depending what you want out of your PC. If this refresh was more like 4.3-4.6 I'm sure a lot of people would sacrifice clock speed for two extra cores providing fast memory is also easy to run. Still seems Intel is a safer bet unless you do heavy workloads, are on a budget or heavily multitask/stream.
Click to expand...

Intel have a lot of problems to work through and the socket platforms are a mess. It's hard (impossible?) to look past a Ryzen based system just now.
 
Clock speed isn't the be all and end all, i would have thought people would have learned that from Bulldozer.

What people are sacrificing in clock speed going from older gen Intel to Ryzen they are more than making up for in IPC, i went from a 4.5Ghz 4690K to a 3.9Ghz Ryzen 1600 and on every occasion the latter is at least as fast, more often than not much faster, sometimes twice as fast.
 
I see the same thing, its very hard to recommend intel at the mo.
A 8700k does not have athe extra 2 physical or 4 logical cores to be able to use like the 2700 has, ok it has a higher clock speed but that will soon be eaten up in the future and the more multithreaded things get the more you need cores.

Unless you absolutely must have that small FPS different that exists in games, which i cannot really see much need for, your much better off with the 2700
 
sideways14a said:
I see the same thing, its very hard to recommend intel at the mo.
A 8700k does not have athe extra 2 physical or 4 logical cores to be able to use like the 2700 has, ok it has a higher clock speed but that will soon be eaten up in the future and the more multithreaded things get the more you need cores.

Unless you absolutely must have that small FPS different that exists in games, which i cannot really see much need for, your much better off with the 2700
Click to expand...

Not forgetting the upgrade path of AM4. Intel, you'll need new mobo and cpu each time.
 
subbytna said:
Not forgetting the upgrade path of AM4. Intel, you'll need new mobo and cpu each time.
Click to expand...

Might be wrong, but doesn't the new precision boost of the next Ryzens only work on 4xx boards? Also we don't really know whether the new boards will do anything for memory speeds so... I'm not quite as convinced by this AM4 path as I was when I first bought into it xD
 
eddiew said:
Might be wrong, but doesn't the new precision boost of the next Ryzens only work on 4xx boards? Also we don't really know whether the new boards will do anything for memory speeds so... I'm not quite as convinced by this AM4 path as I was when I first bought into it xD
Click to expand...

its nice to know you can drop in a Zen 2 next year or 2020- but seems like that will be a different beast .
can see what happened with intel doing coffeelake- z370 has a rush job - socket did need redoing and not the half baked effort . Seems z390 will use the same socket but will be a full chip feature like H370 and VRM count to match - though i bet it'll cost and arm and a leg!

think if i go to Zen 2 , i wont be using First gen board :) but not everyone would want to fork out more cash etc etc - its nice to have options
 
You must log in or register to reply here.
Back
Top Bottom