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

[Boomstick777]

Dbl post.

 

[TheMorningStar]

You wouldn't recommend to anyone regardless of budget, but it would be a great budget CPU... Make your mind up

Cheap FM2+ Mobo + Athlon X4 880K would be ideal for budget PC. The FM2+ does have USB 3.0 + PCIe 3.0 x16 etc anyway..

Horses for courses.

The lackluster feature support of AMD mobos is an issue for me, not to fussed about CPU performance so long as it's decent but things like x4 M.2 support, sata 3, PCIe3.0, Number of sata ports, board presentation/looks. If these don't improve with Zen I'll have to look at intel, if I can even hold out upgrading that long.

 

[Boomstick777]

The lackluster feature support of AMD mobos is an issue for me, not to fussed about CPU performance so long as it's decent but things like x4 M.2 support, sata 3, PCIe3.0, Number of sata ports, board presentation/looks. If these don't improve with Zen I'll have to look at intel, if I can even hold out upgrading that long.

Yeah we were talking about FM2+ though (Which does have USB 3.0, SATA 3, M.2 and PCIe 3.0 anyway)

Zen is coming with updated chipset, DDR4 etc. Different things.

You still on Phenom II? You have done well holding out this long already imho.

 

[humbug]

The lackluster feature support of AMD mobos is an issue for me, not to fussed about CPU performance so long as it's decent but things like x4 M.2 support, sata 3, PCIe3.0, Number of sata ports, board presentation/looks. If these don't improve with Zen I'll have to look at intel, if I can even hold out upgrading that long.

I have M.2 (i think) and native Sata3 on my 4 year old Sabertooth 990FX, why would you need PCIe3? no GPU needs PCIe3 yet.

 

[TheMorningStar]

Yeah we were talking about FM2+ though (Which does have USB 3.0, SATA 3, M.2 and PCIe 3.0 anyway)

Zen is coming with updated chipset, DDR4 etc. Different things.

You still on Phenom II? You have done well holding out this long already imho.

Yeah, I'm holding out for Zen or trying to.. may not make it. Looking to do an incremental upgrade over the next year and a bit. Need new monitors, peripherals, case, HDDs and finish on the cpu/ram/mobo/gfx. Put a budget of 3000, which I can just about justify given I have been... dormant

I'm about to start redecorating so I can move my system to another part of the room so I can fit a 3 screen eyefinity setup, This gonna be a good project

I have M.2 (i think) and native Sata3 on my 4 year old Sabertooth 990FX, why would you need PCIe3? no GPU needs PCIe3 yet.

Yeah, there are a few nice looking boards but the choice is weak. I never intend to buy anything from Asus again for instance. I don't think that M.2 will be PCIe x4 speeds, probably be an implementation that's no faster than sata3 but I could be wrong.

Idealy, I'd be wanting an MSI Gaming 7 or 9 feature level Mobo, high on looks and features. The fact that the primary PCIe slot keeps getting placed lower on AMD mobos is an annoyance. If you look at the MSI 990FXA Gaming board it's literally on the 3rd expansion slot, that just looks like lazy design.

I intend to keep this next base system for at least 5 years, I intend to run a 2 card solution minimum but 4 is a possibility so PCIe bandwidth is mucho importante. It currently looks like 1440p 144hz eyefinity (3) is what I'll be going for. It would have been 3x4k but I'm already gaming on 60hz, want more, more I tell you! oh and freesync. so many variables today

 

[humbug]

Yeah, I'm holding out for Zen or trying to.. may not make it. Looking to do an incremental upgrade over the next year and a bit. Need new monitors, peripherals, case, HDDs and finish on the cpu/ram/mobo/gfx. Put a budget of 3000, which I can just about justify given I have been... dormant

I'm about to start redecorating so I can move my system to another part of the room so I can fit a 3 screen eyefinity setup, This gonna be a good project




Yeah, there are a few nice looking boards but the choice is weak. I never intend to buy anything from Asus again for instance. I don't think that M.2 will be PCIe x4 speeds, probably be an implementation that's no faster than sata3 but I could be wrong.

Idealy, I'd be wanting an MSI Gaming 7 or 9 feature level Mobo, high on looks and features. The fact that the primary PCIe slot keeps getting placed lower on AMD mobos is an annoyance. If you look at the MSI 990FXA Gaming board it's literally on the 3rd expansion slot, that just looks like lazy design.

I intend to keep this next base system for at least 5 years, I intend to run a 2 card solution minimum but 4 is a possibility so PCIe bandwidth is mucho importante. It currently looks like 1440p 144hz eyefinity (3) is what I'll be going for. It would have been 3x4k but I'm already gaming on 60hz, want more, more I tell you! oh and freesync. so many variables today

Why no Asus? i'm curious is all

 

[TheMorningStar]

Why no Asus? i'm curious is all

Awful RMA for years, to many bad experiences, this Phenom II build I didn't want Asus but had to swop to it as the mobo they sold me said it was in stock but wasnt, think it was a DFI board, or whoever it was, they no longer make mobos. would have been a few weeks until they got it back in so the M3A was the closest thing. that died a while back actually using an M4A now, cheep fix from the MM.

 

[humbug]

Awful RMA for years, to many bad experiences, this Phenom II build I didn't want Asus but had to swop to it as the mobo they sold me said it was in stock but wasnt, think it was a DFI board, or whoever it was, they no longer make mobos. would have been a few weeks until they got it back in so the M3A was the closest thing. that died a while back actually using an M4A now, cheep fix from the MM.

Yeah i have read some horror stories about Asus RMA, including one guy who was refused an RMA on a board because of damage, apparent damage that could not be seen in the pix they sent him, they sent him the same board back (without the original box the customer sent it in) and the damage could only be seen under a magnifying glass and they were surface scratches, everyday ware and tear that would not affect the operation of the board, he posted all the pix and correspondence so there was no doubt.

i have to say i really do like Asus boards, i would go as far as to say IMO they make the best boards full stop.

But their RMA and general service is well known to be utterly diabolical.

I might try a Gigabyte board for my next build, service and RMA matters.

 

[TheMorningStar]

Yeah i have read some horror stories about Asus RMA, including one guy who was refused an RMA on a board because of damage, apparent damage that could not be seen in the pix they sent him, they sent him the same board back (without the original box the customer sent it in) and the damage could only be seen under a magnifying glass and they were surface scratches, everyday ware and tear that would not affect the operation of the board, he posted all the pix and correspondence so there was no doubt.

i have to say i really do like Asus boards, i would go as far as to say IMO they make the best boards full stop.

But their RMA and general service is well known to be utterly diabolical.

I might try a Gigabyte board for my next build, service and RMA matters.

I've just taken Gigabyte off my list due to an RMA story a while back where they wanted the user to pay all postage or something like that so MSI is getting my money for gfx cards and mobo most likely unless they start pulling similar tricks. I reward good service with my hard earned money and I expect a 3 year warranty to be a 3 year warranty, my experience with Asus however tells me they only have a 1 year warranty and that's only with the retailer you purchased it from, messing me around for months with gfx card replacements, asking me for codes that don't exist. They must literally owe me hundreds in broken hardware. I really can't see me ever using them again, even if people start raving about how good the care is, that company has literally stolen money from me as far as I'm concerned. won't build a system for anyone if they insist upon anything Asus because I know I'd be handling the RMA, screw that quite frankly, I'm annoyed now thinking about it. I'm going to stop thinking about it now.

 

[Sumanji]

The problem with "blacklisting" manufacturers is that there aren't that many of them. If you take a large enough sample size, they are probably all equally awful, so...

Back on topic; this is hella bad for AMD if true. I thought I read earlier that they were aiming to get Zen APUs onto the market before the performance CPUs? If the APUs won't hit until Q4 '16, we could be looking at 2017 (!!) for the CPUs.

I can see AMD getting acquired / split up before that happens

 

[Orangey]

It's the other way round, CPUs hit first. I have heard server vendors will get theirs before desktop so perhaps late-mid 2016?, if no more delays occur.

 

[Sumanji]

Yeah server first would make the most sense; get competitive high margin parts out asap. I can't remember that far back, but did they do the same thing with Opteron / Athlon 64?

But presumably, if it's a process problem, then no one will be getting anything before Q4?

I really want AMD to do well... not just because I'm a slight fanboy we need some damn COMPETITION in the market!!

 

[dreamcat4]

I'm entirely 'meh' about this whole item. The crux of the problem now seems to have a whole lot to do with expectations. Almost regardless of anything else. It would probably be better for everybody to see AMD hard closing the gap a bit more each generation in some regular and consistent fashion rather than some perpetual promise of things that may be too far out of their reach given the current situation.

People often cite AMD because it's cheaper. But that's not actually the issue. The issue has always been perf/watt in the semiconductor industry. It has been like that since forever.

I think AMD took the right *general* approach (and despite lots of smaller screw ups) with their recent Fury GPU releases. It was not stellar performance, but was a modest improvement.

Now that there's no way AMD can use Zen or any other single-generation to recover their business. This means that needing a single generation 'to recover things' is no longer so important. So a more solid but less-results performance would end up being more credible to outside investment and is still going in the right direction. And that is the more important kind of turn-around which a potential investor needs to see from them at this stage. Not some hype surrounding un-obtainable goals.

Also:

Relying upon a node shrink for new gains is actually a sign of weakness, not strength. That is the perfect way for any manufacturer (be it AMD, Intel, Apple, or anybody else) to covered up a lack of progress / architectural innovation. So to me, I feel that Zen doesn't really mean anything exciting yet in the arena of technical innovation. And morever I probably don't want it to.

After Zen, then AMD will of course be stuck on 16nm for a very long time to come. Which of course is the same boat just like everybody else (who isn't Intel) as 28nm last time around. So during that next era what will AMD be able to do differently if they didn't turn a new leaf we didn't see? Perhaps a lack of public scrutiny is just what AMD needs to come up with some new tricks. Whilst at the same time being scrutinized far more heavily by private investment / outside management.

 

[Nelly]

Saw this posted on overclock.net

Amid challenges, chipmaker AMD sees a way forward

Company, which employs 1,600 in Austin, is counting on its new Zen microprocessor design project to turn its fortunes around.

For more than two decades, Advanced Micro Devices Inc. has been the comeback kid of computer-related electronics.

The chipmaker has worked itself into difficult places only to make Houdini-like escape through clever and pioneering products, acquisitions, business restructuring and through winning legal settlements. Now AMD finds itself in another difficult place — and company leaders think they know the name of the potential escape hatch. It is called Zen — a microprocessor design project whose results are expected to start showing up in a family of new products starting in 2017.

“Everything is riding on Zen,” said analyst Nathan Brookwood with Insight 64. “They are shooting for performance parity with where (arch-rival) Intel will be. AMD understands that they have to succeed with Zen. If Zen fizzles, they will really have to do a lot of running around.”

Suzanne Plummer, the veteran Austin chip engineer who heads the Zen team, exudes confidence about the project.

“It is the first time in a very long time that we engineers have been given the total freedom to build a processor from scratch and do the best we can do,” Plummer said. “It is a multi-year project with a really large team. It’s like a marathon effort with some sprints in the middle. The team is working very hard, but they can see the finish line. I guarantee that it will deliver a huge improvement in performance and (low) power consumption over the previous generation.”

Plummer has worked at AMD since 2002, when it acquired the startup she was working for, Alchemy Semiconductor.

Full read here >> http://www.mystatesman.com/news/business/amid-challenges-chipmnaker-amd-sees-a-way-forward/nngdf

 

[Nutella33]

They are high if they think they can delay Zen to 2017. I hope that's a misprint.

 

[FoxEye]

They are high if they think they can delay Zen to 2017. I hope that's a misprint.

If it's 2017 my next build won't be an AMD. Will be upgrading next year come what may.

I don't think it's a mis-print - 2017 is mentioned twice, and 2016 not at all. So either they're wrong or Zen for desktops is 18 months away.

 

[Orangey]

Oct 2016.

They recently said "first full year of revenue 2017" so this local paper writer probably read that and got confused.

Unless you want to use a random local newspaper writer as an authoratative source above company execs. Can't you just use your own common sense for once?

 

[Poneros]

I think all generational leaps will happen in Winter next year anyway. CPU, GPU, etc. Do a few months really matter that much? Probably not.

 

[Dave2150]

The problem is that Intel will release Kabylake early 2016, followed by 10nm cannonlake in 2017.

Zen will compete with cannonlake, not Skylake.

 

[Sylver123]

The problem is that Intel will release Kabylake early 2016, followed by 10nm cannonlake in 2017.

Zen will compete with cannonlake, not Skylake.

Only at the mainstream. If Intel don`t have Skylake-E out the same time Zen is released then at the high end, Zen will be up against Broadwell-E.

Zen will be an 8 core and Intel mainstream processors are 4 cores so obviously Zen will be up against whatever high end 8 core chip Intel
has out then which will be either Broadwell-E or Skylake-E because cannonlake will be for the mainstream market and have only 4 cores.