Discussion AMD Ryzen MegaThread! FAQ and Resources

Page 21 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.


I haven't read anything about Intel problems with the 10nm node, but I can support you with GloFo skipping directly to whatever they call "7nm". A quick search will probably take you to GFs main page with the information to confirm it.

In regards to the performance numbers being leaked, I think we should really put them to rest now... It's like watching a horse race and having 3 horses running neck to neck here. Anything can be leaked now, so just sit and enjoy the time waiting until the race is over in my humble opinion 😛

I fully back Juan here: wait for the reviews and heat up the pop corn for the 2nd of March (which seems to be the NDA lift date it seems?). I know I have my coffee and pop corn already heating up! Haha.

Cheers!
 
That NDA must run up to the 28th.. As in it expires on 28th at twelve midnight. I think it might expire tonight (with a bit of luck).. So say 12 EST time is 5.00 AM GMT.

I'm sure I read that Lisa Su said it expired on on the 28th for reviews...
 


:popcorn:

Well i was about to say that your missing the popcorn smiley... Then I seen that it's missing... (the name for it is still there but the image is gone 🙁

Well google can always find what your looking for! (identical image from what I remember).

default_popcorn.gif



 


I disagree, I think Global Foundries is doing great work these days. I interned at the facility in Malta, NY. The lab was renovated after they bought the IBM IP with multi-billions.

http://saratogatodaynewspaper.com/index.php/today-in-saratoga/business/item/203-globalfoundries-ups-malta-investment-by-another-$2-billion

And worked with the University at Albany to start this massive R&D project,

http://www.globalfoundries.com/newsroom/press-releases/2016/02/09/suny-poly-and-globalfoundries-announce-new-$500m-r-d-program-in-albany-to-accelerate-next-generation-chip-technology

They even made it to the table first with a 5G RFSOI,

http://www.globalfoundries.com/newsroom/press-releases/2017/02/17/globalfoundries-announces-availability-of-45nm-rf-soi-to-advance-5g-mobile-communications

GF is top-tier these days. I helped set up the 10nm fab line (Which they shut down a while back, and skipped ahead to fitting the 7nm fab line), and it produced really high quality samples. If they say they can manufacture a viable 7nm AMD CPU, I believe them.
 


Don't get me wrong, I'm not saying they aren't doing great work these days (they are), but saying that they will be ahead of Intel(I don't mean nm, I mean quality of production) seems like a bit of a stretch to me. Definite salt on that one. In my humble opinion, their best case is GF 7nm = INT 10nm, or somewhere thereabouts. Also, if you don't mind my asking, how did you come about the internship? Was it part of school?
 


Im getting this nagging feeling that something is going on with Ryzen about now..... A lot of tech youtubers I watch have been strangely quiet today.

So far linustechtips stands out the most to me as he typically uploads a video around 4-5 PM EST everyday (excluding Friday's for his wan show) and it's 7 here now with no video up yet.... Very strange...

EDIT:

Gosh darn it! He published the video while I was typing this!
 


I don't know, I've seen places quoting the 28th and places quoting the 2nd. Not sure when the NDA lifts for actual reviews. Now if I was a conspiracy theory type I would start thinking that maybe the reviews are being delayed because of Intel calling in favors from the reviewers and they are having a sit down chat with Intel before releasing their reviews..... IF I was a conspiracy type person that is😛
 
I don't think it's a conspiracy mate... well not yet anyway..

@ wamon6 I think you could be onto something there.. with a bit of luck they are all busy finishing off their Ryzen reviews !!! Please please lord let it be true...
 


Same thinking we will have to wait to the 2nd 🙁

I'm actually waiting for 2 products so i'm kind of going nuts ha ha
 


I was just making a joke😀
 


Really hope you are right i'll be on toms during that time to just waiting and we can discuss results in the forums here
 
You know with them falling behind in their fabrication process and abandoning their tick tock schedule.. It's clear that they were falling behind, I mean they must have had some attitude of complacency.. They really should have started to develop a new uarch sooner..

I guess they never expected AMD to catch up them.. While now it looks like they are going to pass them out.. on a bigger process node no less.

They really have been caught with their pants down.

Also it looks like TSMC is now ahead of them in Fabrication with their 10nm on schedule for Q1 2017.. As far as I know TSMC's 10nm is a true 10nm.. correct me if I'm wrong here. Here's an articel on TSMC:

http://www.eetimes.com/document.asp?doc_id=1331059

An FYI amd can now avail of either of these new process nodes under their new WSA agreement with GF.

I think they're in trouble with zen; big trouble...
 


IF Ryzen performs the way we are expecting / hoping it to. If Ryzen is what all the early benchmarks suggest it to be capable of then yes, AMD has done the impossible and caught up to Intel's 8th generation iCore. Even if that proves to be true though, don't count Intel out or even assume they where caught totally off guard. Intel has 50x the R&D budget that AMD has and could be sitting on a uarch that will make iCore look as dated as Pentium 4. That kind of money buys incredibly good engineers, I doubt they have all just been sitting around playing darts and drinking beer, although that would be an awesome work environment😀
 
In regards to fabs, nodes are getting really hard to do. It's not shocking 10nm is taking longer then previous nodes. Its going to be a thing going forward. Get used to being on the same node for over two years; it's going to be normal going forward.
 


Well it is evident that they have fell behind in Fabrication... I do hear what your saying an I do agree with you... I was worried myself that they may have been holding back technology. Although on looking into it their roadmap states there next new uarch is a ways off... now if they amazingly just pull one out of a hat we will know they were indeed holding back tech an milking us in the meantime.,. I wouldn't rule it out, but I think it's unlikely, I think it's more likely that they got complacent over the years an never taught AMD could catch them...

An by the way I'm sure there is beers in the fridge (not in Intel Ireland as that would be a really bad idea) but in states yeah as far as I know, plus games room, pool table an probly a gymnasium. I'd say the engineers do work as well as play though...Hewlett Packard in Holland keep's beers in the canteen for there staff (although they do not do that here in Ireland either for the same reason)...

Samsungs 10nm Finfet also available this spring and AMD can also avail of this if they want.. this may be a better choice as it is very similiar to the process their using now:
http://www.anandtech.com/show/10765/samsung-10nm-massproduction
 
Found this on Anandtech... :-(

"Yeah a youtube video with the second part of the Ryzen presentation (was taken down shortly after) showed the AMD brass on stage confirming the Embargo ending on the 2nd (Bummer)."

https://forums.anandtech.com/threads/so-which-is-nda-lift-date-for-ryzen.2500379/
 


Yeah was looking at anandtech twitter page (well Ian Cutress twitter technically) and seen similar thing.

Although... That image in his twitter shows March 2, 2017 9:00 AM EST for the article...

So i think we have our exact time now 😀

C5v9ZSZWcAEvWtq.jpg:large
 


Welp, i should have known better why all the tech youtubers i watch went quiet.... Something something nvidia..... something something ti.....
:lol:

Don't know why it didn't click in my head even though I knew nvidia had their event going on today........
 


it would not make economical sense for intel to develop a far superior product, hold on to it, then release it when/if AMD catches up. They already did the expensive part of developing it. Thats a sunk cost. "milking" the current architecture just puts off releasing something that could make 20% of people upgrade vs 5%.
 
Sorry i use ps4 to browse web so cant post links.
This has been debated at many tech forums and the general consensus is glofo 7nm is about equal in density to intels 10nm, if you look at ryzens power efficiency compared to intel 14nm vs 14nm ryzen is more efficient and smaller, now for the record im not suggesting samsungs 14nm lpp is better than intels 14nm+, far from it, many of those advantages are due ryzen being a well designed newer uarch (no 256b Fmac) and very dense design, its alomost certain that if ryzen was built on intels mature 14nm+ process it would be even smaller and better performing that it is on 14nm lpp.
So despite that 14nm lpp is not that far behind intel as you would think, especially at lower frequency, intels process advantages probably are the most whilst pushing 3ghz+.

From early projections from both companies of improvements to their next node, 10nm and 7nm respectively, it brings them about level.
Please note glofo 7nm is a ground up design using IBM engineering and ip (with glofo) when it was purchased with the fabs, this is not a relashion to samsungs 14nm , it is projected to be MORE than 60% more power efficient and approximately 2x area reduction (thats off of memory). That compares favourably with intels 10nm.

Intels manufacturing lead is shrinking as they have had multiple delays in recent years, Tsmc moves to 10nm this year and then 7nm in 2018, node sizes are not comparable between fabs, its generally assumed tsmc/samsung 10nm ROUGHLY equals intdls 14nm+, at 7nm they will be matching intels 10nm.
It thought that tsmc and glofo will have a better 7nm than Samsungs, both will be in production in 2018.

 
"power efficiency compared to intel 14nm vs 14nm ryzen is more efficient and smaller"

I''ll wait to i see actual power consumption charts and not TDP figures that by now in the eco friendly world we live in they NEED to be using actual power ratings and not heat-output

People can correct me if i'm wrong but isn't that all they mean? TDP measured in heatoutput not actual wattage figures?

This goes at Intel and Amd not just a knock on Amd my 4790K at stock used 15-20 watts more then what intel claimed when maxed out.
 
@jdwii, TDP stands for 'Thermal Design Power' so yeah. It's inherently not an electrical power dissipation measurement as its intended to help builders choose the correct thermal solution. People tend to confuse the two as electrical power dissipation does directly translate to thermal, they are linked, however neither company actually rate their parts like that...