News AMD Ryzen 7 5800X Emerges As A Serious Rival For The Intel Core i9-10900K

Endymio

Reputable
BANNED
Aug 3, 2020
725
264
5,270
>> "The chipmaker's mainstream processors and APUs are currently sporting the Ryzen 3000 and Ryzen 4000 monikers. The current mismatch lends to confusion, especially for the uninitiated ..."

The "uninitiated" don't care what generation a processor belongs to, just what the overall performance will be.
 
  • Like
Reactions: P1nky and Soaptrail

GeekyOne

Reputable
Oct 24, 2019
34
6
4,535
So are they going to release 5000G APUs? I still can't get a 4800G and am waiting for a fast APU to save close to $200 (for a discrete GPU I don't need) since I'm not a gamer.
 
  • Like
Reactions: digitalgriffin

eklipz330

Distinguished
Jul 7, 2008
3,033
19
20,795
>> "The chipmaker's mainstream processors and APUs are currently sporting the Ryzen 3000 and Ryzen 4000 monikers. The current mismatch lends to confusion, especially for the uninitiated ..."

The "uninitiated" don't care what generation a processor belongs to, just what the overall performance will be.
are you kidding or are you oblivious? the uninitiated will just buy whatever is popular and has the highest number in the product name.

that's why intel purposely makes their nomenclature confusing. they can upsell garbage.
 

hannibal

Distinguished
So are they going to release 5000G APUs? I still can't get a 4800G and am waiting for a fast APU to save close to $200 (for a discrete GPU I don't need) since I'm not a gamer.

That is the Main point of being worried. I am sure that Zen3 is gonna be fast. But can TSMC produce enough of those... I doupt very much. There is huge shortage of 4000 series mobile chips and 4000 series apus for AIB parners.
The AMD demand and TSMC production capasity Are in huge imbalance at this moment and very good Zen3 does not improve that situation... Where the h*** amd manage to get production capasity to GPUs in this situation, where They don`t even have enough to their cpus.

That is interesting situation. What prices retailers will put on those if They Are rare as a gold?
 
Interesting results...!

(I look forward to seeing a variety of popular games' results, and not potentially just a cherry picked sample)

If this alleged 5800X can even match the 10700K in most games, and at even a $50-$100 savings, it will be the first time in a VERY long time (the early Pentium IV days) that AMD has outperformed Intel in gaming. Having long since been curb-stomped in Cinebench, to then have AMD match/exceed the 10900K's gaming performance ,a nd with two fewer cores, would be....well.... brutal for Intel...
 

Endymio

Reputable
BANNED
Aug 3, 2020
725
264
5,270
So are they going to release 5000G APUs? I still can't get a 4800G and am waiting for a fast APU to save close to $200 (for a discrete GPU I don't need)
If you don't need 3D graphics, there are plenty of discrete graphics cards in the $25-$75 range, with more than enough performance to handle 2D tasks. For instance:

EVGA GT710

are you kidding or are you oblivious? the uninitiated will just buy whatever is popular and has the highest number in the product name. that's why intel purposely makes their nomenclature confusing. they can upsell garbage.
I've had enough of argumentative obstreperous fanboys today. So I'll simply agree: every Intel chip ever made is utter garbage, and anyone who ever bought one is an unenlightened savage, incapable of performing even the most basic of daily tasks without constant supervision.
 

Endymio

Reputable
BANNED
Aug 3, 2020
725
264
5,270
That is the Main point of being worried. I am sure that Zen3 is gonna be fast. But can TSMC produce enough of those... I doupt very much. There is huge shortage of 4000 series mobile chips and 4000 series apus for AIB parners.
Even many of the 3000 series have escalated rapidly in price, particularly on the lower end. The 3600, for instance, has risen from $155 to $205, in little over a month.
 

GetSmart

Commendable
Jun 17, 2019
173
44
1,610
Compared to someone else's results (source https://www.ashesofthesingularity.c...k-result/69ccbd07-baa6-45b2-9598-b0e6bd81d1d7 )... :unsure:
50398047407_0dcd9be04a_o.png

And this one have same amount of 32GB memory.. :vip:
 
Last edited:

Shadowclash10

Prominent
May 3, 2020
184
46
610
Why, oh why did another Intel v. AMD flame war start when we were just talking about the naming scheme. Seriously people?

I do find it annoying that they skipped the 4000 series except in what, mobile?
 

tummybunny

Reputable
Apr 24, 2020
33
31
4,560
'On one end, the chip could belong to AMD's upcoming Zen 3 (codename Vermeer) family, which the chipmaker will present on October 4.'

But later on October 8 is mentioned, which is the date I've been looking forward to. What happens on Oct 4?
 
Even many of the 3000 series have escalated rapidly in price, particularly on the lower end. The 3600, for instance, has risen from $155 to $205, in little over a month.
That could very well be due to AMD cutting production as they focus on manufacturing the new models of CPUs and graphics cards. Retailers are probably raising prices to avoid running out of stock before the new generation of processors arrive.

The 3600 was never regularly $155 though. Going by PCPartPicker's price history graphs, it only saw that price at one retailer for a brief 3-day sale, and that was over two months ago. For the most part, it had been priced around $175 at major online stores for most of this year, with some occasional dips a bit lower.

At around $205 or more it doesn't make much sense when the 3600X with slightly better clocks and a better stock cooler can currently be had for as little as $210. If one doesn't immediately need a processor though, it's probably worth waiting to see what AMD announces next week.
 

InvalidError

Titan
Moderator
This is the one I've been waiting for and while these "leaks" look promising, I'd rather wait for actually confirmed tests seeing stuff of this sort can easily be manipulated or straight up faked.
Regardless of leaks, there is no doubting that flattening the CPU interconnect hierarchy by having 8 cores and 32MB of L3 cache per CCX instead of 4C/16MB x 2 CCX per CCD will help considerably with performance consistency, especially in situations where threads would have needed to communicate across CCXes on Zen 1&2 but will be on the same one on Zen 3..

Core-to-core latency across CCXes (or even worse, between CCDs) is likely one of the single biggest impediments to Zen 2 beating Intel in most games.
 
Regardless of leaks, there is no doubting that flattening the CPU interconnect hierarchy by having 8 cores and 32MB of L3 cache per CCX instead of 4C/16MB x 2 CCX per CCD will help considerably with performance consistency, especially in situations where threads would have needed to communicate across CCXes on Zen 1&2 but will be on the same one on Zen 3..

Core-to-core latency across CCXes (or even worse, between CCDs) is likely one of the single biggest impediments to Zen 2 beating Intel in most games.

I don't disagree. I'm interested to see how the CCX and IF changes improve performance across the board. Streamlining the process of communication could only help speed things up, but I have to wonder why it's taken this long to make such a dramatic change? One would think that AMD would have known about this prior to Z2 and make the changes then rather than wait until now unless there was something in the architecture itself which prohibited it. I suppose we'll see just how marked an improvement this is soon enough.

No matter what, we're benefiting from it and I can't wait to build a new rig. I've greatly enjoyed my FX8350 and have never felt disappointed but this is going to be a whole new ballgame for me and others like me who have waited to reap the rewards.
 

InvalidError

Titan
Moderator
Streamlining the process of communication could only help speed things up, but I have to wonder why it's taken this long to make such a dramatic change? One would think that AMD would have known about this prior to Z2 and make the changes then rather than wait until now unless there was something in the architecture itself which prohibited it.
There is no miraculous new realization at play here.

The reason why things cannot be scaled arbitrarily large is because larger logic blobs are slower so you cannot scale them up more than the process can offset increased logic blob sizes. If you try scaling them up faster than the process improvements can accommodate, you end up taking disproportionate hit on clocks, die area, power and heat.

Zen 1 had the split CCXes simply because 16/12nm was too slow to put all the cores and L3 in a single CCX. Zen 2 likely retained the split to avoid changing too many variables while AMD was in the process of figuring out the limits of what it can do on TSMC's N7 - if you aren't sure a given feature will meet timings on a new process, better off setting it aside for next-gen than having to a last-minute re-design when test silicon fails to achieve expected performance.
 
  • Like
Reactions: BeedooX
I hope that limit is configurable or not all top chips will run 150W, in SFF 150W cpu will be to much, I would love to ride zen3 cpu with a bit less glow from my ITX box. GPU is taking a lot thermal space I have :(
what do you guys think, R5600 will be 60W or 90W?