*** AMD "Zen" thread (inc AM4/APU discussion) ***

[SimBy]

Well hopefully it is - everybody remembers the TLB the first Phenom had,so lets hope its something not negative!!

Supposedly it's something about memory timings making a large difference. I don't think its any major erratas or deal breakers. Just AMD trying to look as best as possible.

 

[harney]

It will be down to bios update changes

 

[loxocemus]

im on the fence and thats not enough of a drop to get me off it.

Yeah, 7700K £330, down from £360, its still £100 too expensive....

 

[edgedemon]

same, everytime i look at the perfomance increase and the cost of upgrading my i7 920 i go forget that. Now i'm thinking give it till summer for the 6c and everything to bed down and i'll be chomping at the bit to upgrade.

Yeah, I look at it that way as well, 3930K, 32Gb Ram, dual 290X, Rampage extreme mobo and Im struggling for a need to upgrade as Im happy with my speed at the moment. I am thinking of re-benchmarking my rig in an effort to convince myself once official benchies are out though

 

[CAT-THE-FIFTH]

Supposedly it's something about memory timings making a large difference. I don't think its any major erratas or deal breakers. Just AMD trying to look as best as possible.

Well the Phenom II memory controller was also latency sensitive(even more so than pure bandwidth) IIRC.

 

[Frozennova]

Well the Phenom II memory controller was also latency sensitive(even more so than pure bandwidth) IIRC.

It was, I saw quite a big jump in Memory bandwidth using some PSC IC CAS6 1600mhz Mushkin RAM, could only manage around 1780mhz on my 955BE. The same RAM was stable in my Ivybridge rig @ 2222mhz CAS8

 

[awaybreaktoday]

More memory added to supported list..

http://dlcdnet.asus.com/pub/ASUS/mb...IR-VI-HERO_DRAM_QVL_forAMDRyzenProcessors.pdf

 

[CAT-THE-FIFTH]

OK,this was posted over on OcUK regarding the "issue" with Jayztwocents:

OrangeKhrush[/URL]]I accidentally got him in trouble, but was not fully liable, the mod of the forum we discussed forgot to make the link untraceable and it was traced to source, he got a reprimand from AMD. Basically his information to me is me asking questions IE: if my calculation is correct, Ryzen at 3.7Ghz scores 1602 and at 4.1Ghz scores ~1750 give and take .15% margin of error depending on setup. and he would respond by say, I can confirm that is 100% correct.

He has had to correct information regularly but that is due to AMD firmware updates, initial samples had horrible turbo boost issues to the point it was just more sane to turn off. Lets say I trust him, and thankfully he didn't get removed from AMD's testing list and thankfully he has been able to help sort the fake from teh real. I think there may be one ore two secret surprises to come, but we can accept it is far and beyond what Bulldozer was. In fact I have practically forgotten about that AMD.

I have somewhat of a fancrush on Lisa Su, I will admit I didn't like the decision at first but I got it wrong

 

[CAT-THE-FIFTH]

Some more info from SA forums(I think the chap is AdoredTV):

There's a lot of issues affecting results right now. SMT harms performance in some games, BIOS in general is really bad but being updated daily. Windows also needs to be set to "high performance" power mode. As Fott just aluded to, you can relax about memory performance.

It's just the kind of stuff you might expect with a new arch, but makes me wonder just how much of a monster Ryzen will be after all the issues are ironed out.

Any weird results you see are likely down to the above issues so keep that in mind, especially as we only found out about much of it a couple of days ago. I still don't have my sample due to a bunch of unlikely mishaps and I won't be getting it in time for the reviews tomorrow now. Bit of a shame as I'd already put a crapload of work into this, new equipment etc, but ************ happens.

 

[ben90]

Digital foundry and Anandtech, hardware unboxed are good too.

I like oc3d reviews. They are long but very often he has little things that nobody mentions about.

Paul from Paul's hardware is honest, quite a down to earth guy. For a basic overview and results comparison he can be trusted

Hexus is not too bad and they are quite open to feedback if people feel they have missed something. Its UK based like Bit-tech which is owned by the same person/company.

Personally I'll be looking at Guru3D and AnandTech first, then I'll probably have a skim of Hexus and TomsHardware reviews after.

Thanks everyone. Will save this post for tomorrow; should give anyone a decent round up.

 

[CAT-THE-FIFTH]

So the bug must be SMT causing a performance regression in certain games.

 

[Panos]

Sisoft entry for a 1700 @ 4.1ghz

http://ranker.sisoftware.net/show_run.php?q=c2ffc9ef8eefd2ead2e7d0e3c5b78aba9cf99ca191b7c4f9c1&l=en

Interesting. The 1700 with the 65W TDP can do 4.1, possibly on air (doubt many reviewers will watercool).
The 1800X at 95W how far can be pushed? Hence my post from yesterday. The 5.2Ghz on LN2 was very low, just for the sake to break the record of the 5960X @ 6ghz and not give 10 days to Intel to reply with big price cuts.

 

[haszek]

Interesting. The 1700 with the 65W TDP can do 4.1, possibly on air (doubt many reviewers will watercool).
The 1800X at 95W how far can be pushed? Hence my post from yesterday. The 5.2Ghz on LN2 was very low, just for the sake to break the record of the 5960X @ 6ghz and not give 10 days to Intel to reply with big price cuts.

it's 65W at stock, when you oc it's more so it doesn't matter. 1800x is rated higher as it's got base higher and boosts higher

 

[Crockett]

Thanks for all the review recommendations as well. I usually look at Toms for written review and then places like PCPer and random YouTube reviews after.

 

[Deleted member 66701]

Interesting. The 1700 with the 65W TDP can do 4.1, possibly on air (doubt many reviewers will watercool).
The 1800X at 95W how far can be pushed? Hence my post from yesterday. The 5.2Ghz on LN2 was very low, just for the sake to break the record of the 5960X @ 6ghz and not give 10 days to Intel to reply with big price cuts.

I dunno what the rankings really mean, but Ryzen ranked first?

http://ranker.sisoftware.net/top_system.php?q=c0a598a88efdc0f0&l=en

 

[iHATEMacs]

You are not new, you should know by now that Loadsa$ is almost always jokingly trolling

I'm new, what's trolling? Is it like 'Troll On Floor Laughing'? (TOFL)

 

[SiDeards73]

Interesting. The 1700 with the 65W TDP can do 4.1, possibly on air (doubt many reviewers will watercool).
The 1800X at 95W how far can be pushed? Hence my post from yesterday. The 5.2Ghz on LN2 was very low, just for the sake to break the record of the 5960X @ 6ghz and not give 10 days to Intel to reply with big price cuts.

What's also interesting is all that oc'ing was done on immature bios etc affecting memory performance and possibly overall performance.

Will be interesting to see how these chips are on a matured platform

 

[drunkenmaster]

So the bug must be SMT causing a performance regression in certain games.

It's not really a bug there are afaik some games that Intel still have worse performance with HT on and there is more than one professional application where are better off turning HT off.

HT is what it is, it can help push more data through each core, but you're still taking some resources away from the main threads. HT is epic in some loads, in other loads it's much less good, in others still it can even cause a problem in stalling the main thread even now and then and hurting performance, that shouldn't be too surprising.

Intel after a few iterations got it to the point that for end users it was more often a benefit being on than being off that it became the home user default to always have it on.

Zen will be really damn good, but the first iteration in general leaves the most to be improved, Zen+ could easily have a significantly larger boost than Intel's next maybe even couple of iterations put together. Zen+ could potentially put AMD right on par with Intel IPC.

 

[drunkenmaster]

What's also interesting is all that oc'ing was done on immature bios etc affecting memory performance and possibly overall performance.

Will be interesting to see how these chips are on a matured platform

I think we're back to the revision B mobos tbh. Abit 2.0... I remember that bit but can't actually remember what board or chipset that was
With a completely new platform the first motherboards are a little bit of a shot in the dark. I expect there to be a couple mobos out of all of them that did something really stupid and just suck at any overclocking, a couple will be epic and the rest will probably improve significantly just through bios upgrades. The worst mobos might get fairly quick(2-3 month) b revisions, the rest will probably hang on till Zen+ which will probably bring an updated range of motherboards with lots more tweaks and lots more ideas about how to get the best out of Zen included in the design.

EDIT:- I just realised, I was referring to the 2.0 in the motherboard, there being a major second version of a board everyone loved, after I posted I realised I hadn't even really noticed I said Abit, that alone is a blast from the past.

 

[Spyhop]

All I hope is that the BIOSes are not so immature that there are stability issues at stock speeds. That could turn into a PR nightmare and rightly so.