Associate
- Joined
- 27 Apr 2007
- Posts
- 963
If you look at their DAW test results for the 2700x you will see that on average the 8700k is 25% faster.
-
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.
Intel’s X-Series Processors Explained - by Mike Jennings
- Thread starter Connor@Overclockers
- Start date
/plays Air Guitar Bill 'n' Ted styleeeeeeee!
This is due to a not optimised code - there are many other applications where Ryzen 7 2700 beats i7 8700k with 30-40% and more.
A single buggy application cannot be a reason not to take the vastly superior CPU 2700X.
Well actually it might be due to latency, which has always been a disadvantage for Ryzen, hence improving it was a focus of Ryzen 2.
Come on ....face facts
Soldato
Poor execution on non Intel chips should not be a surprise for niche/specific/specialist software tbh, AMD were out of the game for a fair few years and still have a small market share - it will take years of decent sales, not to mention development time before some even see it as worth considering - and who can blame them!
Caporegime
No one even knew what DAW was until you started banging on about how Intel are better at it, the fact is you had to go to something that obscure to detract for the rest of Intel's otherwise junk CPU's.
Eh?
If you want performance in games you go Intel aswell...it's not hard to read reviews...
Lol
Deluded....
Not exactly. Was about to post to remind you the fraud with the in-built in CPU-Z benchmark how magically the advantage of Ryzen 7 1700 over i7 2600K went from 3.2-3.3 times http://www.guru3d.com/articles_pages/amd_ryzen_7_1700_review,9.html to miserable 2.3 times. And it happened after AMD released its CPUs and someone got scared and changed the benchmark.
Last edited:
Caporegime
A bit, in cherry picked games at 1080P with a 1080TI.
Its an over priced toaster.
Yeah that's right i'm getting bored of Intel's commercial arm...
Yawn...
It's boring now...The Ryzen brigade are a funny lot
He is right. At high resolutions, the games become GPU-bound. No one should take 1080p seriously in 2018, come on.
Caporegime
Even i don't play anything at 1080P, with a 1080P screen, i use DSR with a GTX 1070.
The usage scenario where the Intel is a better buy is an incredibly narrow user base, probably less than 1%, for the 99% its just not a good chip compared with what else you can get from the other team for less money, and as a bonus you get to keep your warranty and keep cool this summer at the same time.
Soldato
I've no idea how that is related to my post - sorry
Intel is better at some stuff, AMD is better at others, for some tasks Intel makes more economical sense, for others it's AMD.
All im saying is that software development life cycles are long so it might take some time before we see AMD chips meeting their potential in some software...and some vendors may not even bother if they don't believe the demand is there.
Am not an AMD or Intel fanboy - just buy the best chip for the job, at a price I can afford, most people would do the same I assume - it's more about the numbers not the brand
Don't be so melodramatic, I knew what DAW was and I'm sure others did too. I dabble in it a bit myself but not enough to care about real time latency or performance. If you did though, it can make a huge difference.
I will tell you - no matter when in the long software development life cycles, Intel's long arm will influence and cripple non-Intel CPUs performance. They do it all the time.
You are saying that time is needed for things to improve, I say that no time will fix anything.
Soldato
Then you would be wrong - I develop software, and run a team in a big company doing just that, I still code despite being the boss, and like many software devs have inherited old code bases that's can't just be simply made multi-threaded, or moved away from instruction sets or libraries they have depended on for years.
For example several years ago, it was decided by my predecessor to adopt some practices that meant our software could not easily be detached from the Intel compiler, this was a good decision at the time, AMD did not have a presence to speak of in our userbase, and it gave us a substantial performance gain without an otherwise costly and time consuming rewrite for a very low cost - but now, thanks in part to AMD (am a big fan btw) we are seeing chips with more and more threads hit the market, this has allowed us to get a business case through to fund the development of what is essentialy a ground up rewrite, this is not a quick process when you have a large install base of users and a lot of 'tech debt' to pay off - we are what, 18 months into the new world of cheap HCC chips?
Big businesses do not have the kind of agility required to make such shifts in what is a relatively short timescale - they are driven by market forces, and AMD are barely a drop in the ocean compared to the number of Intel chips out there in data centres and workstations right now
Gamers, Overclockers and hardware enthusiasts are not in anyway representative of what is out there in use in the real world - I hope this will change, and I hope we have a strong competitor to Intel once more as it can only be good for us all - but your wrong to think it's going to be swift or easy for all the code out there to be optimised for this new world we find ourselves in I'm afraid.
Not just that, but a lot of the IT people won't even consider AMD server chips because all they know is Intel and don't want to take the risk of going with something new.
I don't think AMD will ever overtake Intel but a world without them will be extremely painful.
I don't think AMD will ever overtake Intel but a world without them will be extremely painful.
Soldato
I think we will see a bit of Datacentre resurgence with Epyc through 2019, it's much more economical than the Intel alternatives for certain workloads as you can achieve 2TB of Memory per socket, and can cram much more NVME and GPUs into a relatively cheap box thanks to the large number of lanes for a lower cost - but like you say it will take a bit of time for confidence to be strong enough to see large deployments outside of the hyper scale players.
Epyc does seem to have a slight advantage in the areas of core count and power consumption, especially in the mid range - but the new Dell and HPE servers in this space are still a little hard to get hold of in volume at the moment, and Xeon still smashes it for some workloads (and the cost difference isn't that massive when you look at TCO due to the fact licences for servers and software can make the hardware look cheap anyway!)
Last edited: