New Zen information, AM3+ info, APU presentation, and video card information

Shintai said:
So instead of relate to the substance, you draw the hate card when you disagree. Well done.
Click to expand...

Hate card? Have you ever said let's wait until final benches? Have you ever said okay this looks good but we don't know yet let's wait it out? No! Every single comment I read from is bash bash bash amd, you are always in to proving zen is a failure. I don't play the hate card, but you fail to be objective at times because you try too hard to prove amd has nothing with zen. Anyways we can give you all the substance you will do your best to find a way to say zen sucks. does substance really matter when it's you vs amd? Anyways good talking to you. If zen sucks it sucks you or no one else at the moment have any proof that it's a fail. If it's a fail when it launches then it's a fail.
 
It's a bit worrying that they won't mention any TDP numbers, but I'll wait until it is released before making any serious conclusions. At this point it's just rumors.
 
-Strelok- said:
It's a bit worrying that they won't mention any TDP numbers, but I'll wait until it is released before making any serious conclusions. At this point it's just rumors.
Click to expand...

I think they are shooting for better tdp. From what someone said at the demo they showed that tdp was something they heavily focused on. As you said we will wait and see. I am giving amd benefit of the doubt here because they have had a while To develop the architecture. I don't know about clock speeds though. I hope it tops out at 3.5 at least. At 3.0 it might be held back, but I think they also wanna scale this in to next revision. Just like intel has been doing.
 
I thought it was 95 TDP being the 8/16 part. Or maybe that was speculation but thought that was direct from AMD.
 
cageymaru said:
AMD Zen : The huge architectural leap to Excavator in detail.
AMD Zen: Der riesige Architektur-Sprung zu Excavator im Detail

Translator required, but darn interesting!
Click to expand...
Anmerkung: Wenngleich die Präsentation von AMD erst am späten Dienstagabend Ortszeit in Cupertino über die Bühne geht, erhalten Teilnehmer wie ComputerBase vor Ort bereits am ersten Tag alle Informationen und Präsentationen, komplett ohne eine Sperrfrist. Vor Ort hat aber auch ComputerBase noch einen Termin mit AMD am Dienstag und wird die Meldung bei Bedarf aktualisieren/ergänzen.
Click to expand...

Note: Although the presentation of AMD local time in Cupertino goes across the stage until late on Tuesday, participants will receive as Overclockers spot on the first day all information and presentations, complete without a waiting period. also site Computer Base made an appointment with AMD on Tuesday and the message update / supplement as needed.

More on Tuesday it seems :)
 
Pieter3dnow said:
I like his blog one of the few sources that are worth using regarding Zen.
Click to expand...

Speaking of Dresdenboy, he's made a new post on Zen's die size and I think speculation on Zen's schedulers?

dresdenboy.blogspot.com/2016/08/two-days-to-go-until-amds-hot-chips.html?m=1
 
The bullet item "Two AES units for security" interests me and raises a few questions. Is that per chip? Per core? Other? How many does intel have?
 
Darakian said:
The bullet item "Two AES units for security" interests me and raises a few questions. Is that per chip? Per core? Other? How many does intel have?
Click to expand...
2 for the whole cpu ?
Didn't AMD have the ARM core handle certain things?
 
Pieter3dnow said:
2 for the whole cpu ?
Didn't AMD have the ARM core handle certain things?
Click to expand...
In carrizo yes, but that was effectively just one AES unit (I think). This is just a line item, so who knows what it really means, but they're boasting about it so I guess the standard is one.
 
AES will be of limited use on consumer devices or workstations, but of great use on servers doing lots and lots of encrypted communications.
 
I was just scared that I would have to wait on NowInStock to win the refresh lottery to get a chance at one if they are decent. :)
 
  • Like
Reactions: NKD
like this
lol ya, I hate the bitcoin miners and their furnaces that they build. That said, zen does support a sha1 command and bitcoin does use sha1. That said, intel also supports this and bitcoinists don't seem to go for intel chips for their furnaces, so zen is probably safe. This much AES is much more a server oriented feature for things like TLS connections, encrypted databases, etc.

Edit:
Upon further reading, it looks like AMD is introducing two encrypted memory modes with zen. If those use AES these units might be mostly in service of this new feature.
 
Last edited:
cageymaru said:
Microsoft made 'em do it: The latest Kaby Lake, Zen chips will support only Windows 10.
Microsoft made 'em do it: The latest Kaby Lake, Zen chips will support only Windows 10

I read some months ago that they would work in older OS, but don't expect certain optimizations to be enabled as they are only found in WIn 10.
Click to expand...

Yep, but just wait for the crap storm, illogical, off the wall claims that will arrive with the Front Page News thread. (We know it is coming and we know who will be there, me included. :D )
 
Don't hold your breath for it.
Click to expand...

I assume they care about market share and would rather have customers instead of telling them to buy an Intel processor if they want windows 7.
 
drescherjm said:
I assume they care about market share and would rather have customers instead of telling them to buy an Intel processor if they want windows 7.
Click to expand...
If the article above is anything to go off of then it's a microsoft decision not an AMD decision.
 
However AMD should have the ability to provide a processor driver for its own CPU. That is unless Microsoft tells them they can't.
 
So I'm waiting to build my HTPC until something with better integrated GPU's comes out (I want as few parts as possible). I'm really looking at Zen because AMD has always been better at APU's. But Intel is doing their GT4e GPU's in the next Skylake releases.

Do we think Zen's APUs are going to still be better than Intel's?
 
Compared to a Haswell Xeon with normalized clocks, the ST performance is really low. The Haswell is about 50% faster. (1.6Ghz Xeon as example.).

It also seems a 1.45Ghz Zen fits around a 1.8-2Ghz Pilediver in ST.
 
Shintai said:
Compared to a Haswell Xeon with normalized clocks, the ST performance is really low. The Haswell is about 50% faster. (1.6Ghz Xeon as example.).

It also seems a 1.45Ghz Zen fits around a 1.8-2Ghz Pilediver in ST.
Click to expand...

seems quiet low. I am thinking the es sample has some shit disabled and not clocking at turbo at all or the cache. If what you are saying is true then there is barely no IPC improvement over excavator? I highly highly doubt that. Since Excavator core is suppose to be already faster than piledriver. Something is missing in that benchmark.
 
NKD said:
seems quiet low. I am thinking the es sample has some shit disabled and not clocking at turbo at all or the cache. If what you are saying is true then there is barely no IPC improvement over excavator? I highly highly doubt that. Since Excavator core is suppose to be already faster than piledriver. Something is missing in that benchmark.
Click to expand...

What should be disabled besides turbo? The single core performance is what it is.
 
Shintai said:
L3 wasn't disabled. If you look at other results, it shows no L3 either. This is just how GB works.
Click to expand...


it shows L3 fine for my test. For now I am taking this with a grain of salt. Cuz my 4 cores are beating 64cores. I am like scratching my head. Even at 1.4ghz 62 cores should make up for it. I think the ES sample might be crippled or something.
 
You must log in or register to reply here.
Back
Top