AMD Zen Features Double the Per-core Number Crunching Machinery to Predecessor

[bug]

You're veteran here and for sure you know what you are saying.
Obviously there were improvement and new instruction sets but they were underwhelming in current applications to say the least.

Of course Intel will still have the performance crown for a long time from now on as they have strong Enthusiast parts, but that is not the mainstream market. The mainstream market is formed by the i3, i5 and i7 non E series.

If anyone, can produce a competitive product here then they are in business, and intel in this area was pretty much sleeping over the years. You said it yourself in those long posts, ivy was die shrink, haswell brought very good power consumption, broadwell its a different beast but I'd exclude as it has expensive edram, and its mostly a very expensive GPU with CPU cores, good for laptops ... but not very viable in my opinion from cost perspective for desktop. Lake mostly die shrink ... again a bit underwhelming.

I found an article which compares different generation performance at same clock also.
It doesn't include the latest architecture though, but it is focused on gaming, which is basically the reason for which most of the people buy these processors, otherwise for browsing or office even the atom is good. If you think is too extreme, fine go with an i3.

See here: http://wccftech.com/intel-sandy-bridge-ivy-bridge-haswell-graphics-compared-10-difference-average/

I put them in a table also, and compared similar products:

Lowest details i7 2600K @4.5 i7 4770K @4.5 % increase
Crysis 3 91 93 2%
Black Ops 2 355.4 382.8 8%
Bioshock 243.2 265.9 9%
Battlefield 3 199.8 200 0%
Unigine Heaven 4243 4280 1%
Firestrike 7292 7466 2%

Average: 4%

This increase you will probably get by just shrinking the original sandy without any architecture change. Maybe things will change once the new instructions sets will start to be used ... but if they are not supported by most of the computers in use, they will mostly be scattered optimizations in one app or another.

So yes, if anybody can get sandy level performance, good clocks and reasonable power consumption they are back in business.

Those tests are only looking at graphics. Here's a more complete set: http://www.hardocp.com/article/2015/08/05/intel_skylake_core_i76700k_ipc_overclocking_review/4
You'll see gains in synthetic performance and mostly encoding and rendering. Other things are in the same ballpark as Sandy Bridge.

 

[AsRock]

It takes at least a year to produce a new processor including design from an existing microarchitecture to prototyping to mass production. If AMD put out a processor that beats 6700K (in today's terms) for $300, Intel couldn't respond with a competitive product for a year. Even assuming they already did prototyping and have a design ready for mass production the shelves, AMD could still get a several month long window to grab market share.

Intel's immediate response would probably be cutting prices across the board which likely moves LGA 2011 into mainstream prices. As I said in the post, a shocker from AMD could turn LGA 1151 into a budget socket overnight and LGA 2011 into mainstream.

Not sure if it would take a year for intel, but so if it did i bet all current chips would drop in price to discourage people from buying other than theirs.

 

[medi01]

Guys, back in the Prescott times Intel had no problems more power hungry slower yet more expensive P4's over Athlon64's, why would it have any problems with Zen?

 

[GorbazTheDragon]

So yes, if anybody can get sandy level performance, good clocks and reasonable power consumption they are back in business.

The thing is you can't just consider gaming performance. SKL is a good 20-30% faster in CPU bound applications... In fact, the FX CPUs still make decent gaming machines, as long as you are not pushing for very high framerates in certain games, but the fact that they are now lagging behind intel in overall performance is the killer, especially considering they are 100-200w parts.

SB level performance on a per core basis will be acceptable, assuming the power envelope for a quad core is around the same or lower than SKL, but they will need to be pushing at least 8 core parts and should consider 24-32 pcie lanes for higher end GPU setups.

Also, they need to be considering at least triple channel memory, especially if they want to put the APUs in the same socket, otherwise there will be 0 reason to not go with an intel part and dGPU. That said, intel should also be considering the same if they want to push their iGPU market forward. And this is all on top of the things skylake already brings to the table as far as chipset connectivity

 

[ZeDestructor]

That sounds like Intel in the P4-Pentium D era.

Bulldozewr has been AMD's NetBurst

It would have made sense on 6700K. Even so, we don't see 6 core Xeons on LGA 15## either.

Never gonna happen on the mainstream chip. It's just too good a place to crush the low-end nVidia/AMD cards. Remember, the 6700K is just an unlocked 6700/E3-1275v5 (whenever that comes out), which are both quite popular in business settings (partly because of the quite decent iGPU). It's also the same core as shared with all other quad-core CPUs, with some cache and HT sliced off for the i5 variants.

We are close to the end of the line with IPC improvements for Intel or AMD, the rest will come through process, cache, and instruction set/hardware support. We are close to the end of the line with Silicon in the high performance categories. We can slap more cores in, more sockets, more memory. But the next big thing is either going to be quantum computing, or photon based.

Pretty much

Only thing that AMD will mess up is the way how mainstream and enthusiast is now separated. Because if Intel will have to bump mainstream up quickly, it means they'll have to bump up enthusiast as well, otherwise they'll make them equal which means they'll sell less of the more expensive enthusiast platforms and CPU's. But that's good as far as consumers go, assuming the prices stay low...

Been calling that since the Broadwell rumour days. KBL or CNL is when that will happen I think.

It takes at least a year to produce a new processor including design from an existing microarchitecture to prototyping to mass production. If AMD put out a processor that beats 6700K (in today's terms) for $300, Intel couldn't respond with a competitive product for a year. Even assuming they already did prototyping and have a design ready for mass production the shelves, AMD could still get a several month long window to grab market share.

Intel's immediate response would probably be cutting prices across the board which likely moves LGA 2011 into mainstream prices. As I said in the post, a shocker from AMD could turn LGA 1151 into a budget socket overnight and LGA 2011 into mainstream.

They probably already have 6-core LGA115x variants internally validated (or nearing validation) and ready to go (I know I would with the rumours currently flying). Based on how close the E launch has been with the EP launch, it looks like most of the extra time needed for E/EP/EX chips is validation in multi-CPU and PCIe validation (based on how everything from quads to 18-core chips launch at the same time), both of which are not needed on LGA115x..

Not sure if it would take a year for intel, but so if it did i bet all current chips would drop in price to discourage people from buying other than theirs.

Some reshuffling of the lineup: increased core counts on i7, HT on i5.

Guys, back in the Prescott times Intel had no problems more power hungry slower yet more expensive P4's over Athlon64's, why would it have any problems with Zen?

Intel had to resort to some seriously scummy anti-competitive tactics back then, got sued by AMD, AMD won, and proceeded to wreck Intel the server space with their way better Opterons. Intel seems determined not to get back into a similar situation ever again.

The thing is you can't just consider gaming performance. SKL is a good 20-30% faster in CPU bound applications... In fact, the FX CPUs still make decent gaming machines, as long as you are not pushing for very high framerates in certain games, but the fact that they are now lagging behind intel in overall performance is the killer, especially considering they are 100-200w parts.

SB level performance on a per core basis will be acceptable, assuming the power envelope for a quad core is around the same or lower than SKL, but they will need to be pushing at least 8 core parts and should consider 24-32 pcie lanes for higher end GPU setups.

Also, they need to be considering at least triple channel memory, especially if they want to put the APUs in the same socket, otherwise there will be 0 reason to not go with an intel part and dGPU. That said, intel should also be considering the same if they want to push their iGPU market forward. And this is all on top of the things skylake already brings to the table as far as chipset connectivity

AMD already has quad-channel memory on their high-end Opterons. What they really need to do is move the PCIe controller into the CPU and turn HyperTransport up to 11 in order to make cross-CPU PCIe access not crap in order to match what Intel is doing on LGA2011.

 

[GorbazTheDragon]

AMD already has quad-channel memory on their high-end Opterons. What they really need to do is move the PCIe controller into the CPU and turn HyperTransport up to 11 in order to make cross-CPU PCIe access not crap in order to match what Intel is doing on LGA2011.

I'm pretty sure they will have an integrated PCIe controller, especially since they want to have the same socket for both the normal CPUs and APUs. Opterons have no influence on the consumer market, even less than intels high end Xeons.

HyperTransport is essentially the equivalent to DMI when AMD moves to an integrated northbridge (memory/PCIe controller) so to accommodate the faster chipset interfaces (PCIe3/SATAe/M.2, high sata 6g count, USB3.1, etc) they will NEED to make improvements there.

 

[RichF]

#11

As far as it can be known AMD will release Zen on 14nm (GloFo) or 16nm (TSMC) FinFET technology.

By the way you are right when you say you'd compare Zen to SB. If Zen reaches SB's performance level I would say well done!

Based on my own Cinebench R15 single thread results calculations, SB has app. 45-50% more IPC than Piledriver/Steamroller and ~30% more than K10. (See how bad is the Bulldozer family?) So reaching SB's performance level would be a great leap forward.

Another question is that it'd be still behind Intel's actual performance level.

Cinebench is an Intel-centric benchmark as far as I know because it emphasizes FP.

 

[ZeDestructor]

I'm pretty sure they will have an integrated PCIe controller, especially since they want to have the same socket for both the normal CPUs and APUs. Opterons have no influence on the consumer market, even less than intels high end Xeons.

Err.. wat? Do you not remember the good old S940 Athlon FX chips? Those were straight up binned Opteron chips, down to requiring ECC FB-DIMMs on consumer boards. Single-CPU Opteron platforms also share the same socket as desktop Opterons (AM3+ right now). You need to go up to multi-socket server before Socket C32 and Socket G34 come into play. They also reuse Socket FT3 (BGA relative to AM1) for the APU-based server chips.

HyperTransport is essentially the equivalent to DMI when AMD moves to an integrated northbridge (memory/PCIe controller) so to accommodate the faster chipset interfaces (PCIe3/SATAe/M.2, high sata 6g count, USB3.1, etc) they will NEED to make improvements there.

No. AMD so far has only moved the memory controller to the CPU on the big cores, much like LGA1366 Nehalem/Westmere from Intel. This results in HyperTransport being used to do inter-CPU communication and CPU-NorthBridge communication, with a seperate SouthBridge. HyperTransport is really much closer to QPI than DMI.

Of course, the APUs have been updated much more frequently, and have over time had the PCIe controller integrated into the CPU. Rumours say that they want to integrate the SouthBridge too at some point, which would be quite fun to see.

The DMI equivalent on AMD's side would be A-Link Express or UMI, basically a tweaked PCIe 4x interface.

Cinebench is an Intel-centric benchmark as far as I know because it emphasizes FP.

There's no Intel preference, just that AMD didn't put in as much FP hardware in Bulldozer (based on the bet that basically everyone would port FP code to GPUs). If you compare a 6-core Thuban to a 6-core Bulldozer, The older Thuban core beats the Bulldozer in Cinebench MT.

 

[medi01]

Intel had to resort to some seriously scummy anti-competitive tactics back then, got sued by AMD, AMD won, and proceeded to wreck Intel the server space with their way better Opterons. Intel seems determined not to get back into a similar situation ever again.

Opteron never went beyond 12% of the market share. (despite all the cheaper/cooler/faster)
Intel has not been defeated in court, they've merely settled for 1 billion $. (a laughable sum, compared to the damage done)

There was hardly anything new as far as anti-competitive practices go in times of Prescott, AMD had the same problems prior to that.


Neither did nVidia lose much of the market share, despite Fermi fiasco.


PS
On a side note. Does any AMD CPU beat i5 750 (a 45nm product) IPC wise? Or have something with comparable performance but better perf/watt?

 

[ZeDestructor]

Intel went from 5% to over 25% in a matter of 2 years from K8. By 2007 though, Core 65nm (Conroe) was ready, so that's where it stalled and Intel took the crown back.

Fermi succeeded despite it's issues because of CUDA, even though nVidia lost a solid 10% marketshare (thanks to dbz over from beyond3d). In comparison, AMD's Stream API was hopeless, and OpenCL was just a pipe dream

On the i7-750, the FX-8150 beats it in some places, and loses out in others. An FX-8350 in comparison is either neck and neck or faster

 

[medi01]

AMD went from 12% to 25% (more than I thought) but still rather a modest change, considering that we are talking about product that was better on all fronts.

CUDA argument is laughable.

As a bottom line, we see slower more expensive more power hungry products by AMD competitors only mildly affect the market share (10%-ish swing, if at all) while AMD losing one of the points can lead to it's market share to be cut in half, this looks particularly bad in graphic cards market, where AMD has more than competitive products.
.

 

[R-T-B]

Fermi succeeded despite it's issues because of CUDA, even though nVidia lost a solid 10% marketshare (thanks to dbz over from beyond3d). In comparison, AMD's Stream API was hopeless, and OpenCL was just a pipe dream

Agreed on Stream (it was destined to die), but I think you are massively underestimating the market penetration of OpenCL.

 

[progste]

I hope we finally get a great AMD CPU lineup

 

[Aquinus]

AMD's Stream API was hopeless, and OpenCL was just a pipe dream

OpenCL is not a pipe dream, it's a tool for a very special group of problems. The issue is that OpenCL is only helpful in a handful of situations where mass parallel processing on large data sets is advantageous. You think making applications multi-threaded is hard? Imagine having only parallel compute at your disposal and that's what OpenCL basically is. One doesn't tend to just write any application using OpenCL, it tends to compliment other applications that aren't OpenCL, not stand out on its own. Just wanted to point that out. You use the right tool the right job. You don't go writing real time applications in PHP and you don't to write web applications in C (cgi-bin). You use the right tool for the right job.

What is a pipe dream is AMD thinking that they could make a pipeline as long as Netburst but not encounter the same problems as Netburst. That's what I call a pipe dream.

 

[Arjai]

That's it. I am seriously going to talk with my Broker.

Zen, could make or break my investment future. Or, Not. I figure 1 to 2 Hundred of investment, now, could be a big deal or a minor loss.

Putting it on my Monday Calendar. I am gonna Bank on the new Zen chip. There's just too much to possibly lose, not doing it!

 

[bpgt64]

I would love AMD to take a good look a comparable ShadowPlay esq feature. It's been the only way(outside of a 1,000US investment in a 4k external capture system) to record in 4k for most games.

 

[librin.so.1]

I believe they already have that.

 

[64K]

AMD has two battles to fight. The biggest one isn't whether Zen is a good performer compared to whatever Intel has in the battle chest to counter Zen if it's strong. The biggest battle they have to fight is getting their chips in the Computer Manufacturer's PCs at a decent profit. It does little good for AMD to have great chips if they can't sell them at a decent profit.

http://www.investopedia.com/stock-a...oblems-facing-advanced-micro-devices-amd.aspx

"AMD competes with Intel in PC and server processors but has been bleeding share in both markets. In servers, AMD won as high as 25% market share in 2006, making it a major player in the industry. Today, AMD is essentially nonexistent in the segment, claiming a low single-digit share. Meanwhile, Intel has built a near-monopoly, allowing it to charge high prices and extract extremely high margins.

In PCs, the story is much the same. Intel has continued to steal market share from AMD, especially at the low end with its Atom chips, despite already controlling most of the segment. During the second quarter of 2014, Intel generated nearly 95% of PC processor revenue, shipping 84% of all desktop processors and 88% of all laptop processors."

The PC market continues to shrink as well. I read a report that that may change around 2019 as emerging markets come online but that is just speculation for now.

 

[RichF]

There's no Intel preference, just that AMD didn't put in as much FP hardware in Bulldozer...

There are other benchmarks that don't emphasize FP so much. Just using Cinebench to compare processors is questionable.

 

[alucasa]

I am old enough to know that whatever numbers they are throwing must be taken with a mountain of salt, especially given their past track record.

I will hold my judgement until I actually see this thing rolling.

 

[cadaveca]

I am old enough to know that whatever numbers they are throwing must be taken with a mountain of salt, especially given their past track record.

I will hold my judgement until I actually see this thing rolling.

One might question why this even made it into the public space. Given rumors of sales/merger/breakup, this sort of info is truly only useful to investors or those looking to purchase AMD.

 

[Israsuke]

Good for AMD! So does that mean, Zen won't be an eight-core? More like a quad-cores with large modules? I feel a slight letdown, again good for AMD, if not they will go broke.

I was hoping an eight core that performs what bulldozer should have been from the start. A real eight-core at a decent price. If Zen gets close to the performance of Skylake or cannonlake, I won't be upgrading my 3770k, especially with prices of DDR4. It feels CPU technology has been void and stuck for the last 5 years on performance and speed.

I do a lot of editing and buying an eight-core from Intel feels like a rip off, really a 1000 dollars for a chip? what is it made of pixie dust? Hope AMD gets the market stable on the price, but that also means I'll be waiting at least 2 or 3 years to have a eight-core at a good price, like the rest of us.

 

[ZeDestructor]

Agreed on Stream (it was destined to die), but I think you are massively underestimating the market penetration of OpenCL.

OpenCL is not a pipe dream, it's a tool for a very special group of problems. The issue is that OpenCL is only helpful in a handful of situations where mass parallel processing on large data sets is advantageous. You think making applications multi-threaded is hard? Imagine having only parallel compute at your disposal and that's what OpenCL basically is. One doesn't tend to just write any application using OpenCL, it tends to compliment other applications that aren't OpenCL, not stand out on its own. Just wanted to point that out. You use the right tool the right job. You don't go writing real time applications in PHP and you don't to write web applications in C (cgi-bin). You use the right tool for the right job.

What is a pipe dream is AMD thinking that they could make a pipeline as long as Netburst but not encounter the same problems as Netburst. That's what I call a pipe dream.

At the time OpenCL was a pipe dream. Nowadays the story is very different.

I would love AMD to take a good look a comparable ShadowPlay esq feature. It's been the only way(outside of a 1,000US investment in a 4k external capture system) to record in 4k for most games.

Gaming Evolved

There are other benchmarks that don't emphasize FP so much. Just using Cinebench to compare processors is questionable.

Of course not, that's why I linked two comparisons of a range of benchmarks, where we find that AMD loses where FP is involved (Video/Image processing, Monte Carlo for example), but wins at integer (7-zip file compression for example).

Good for AMD! So does that mean, Zen won't be an eight-core? More like a quad-cores with large modules? I feel a slight letdown, again good for AMD, if not they will go broke.

I was hoping an eight core that performs what bulldozer should have been from the start. A real eight-core at a decent price. If Zen gets close to the performance of Skylake or cannonlake, I won't be upgrading my 3770k, especially with prices of DDR4. It feels CPU technology has been void and stuck for the last 5 years on performance and speed.

I do a lot of editing and buying an eight-core from Intel feels like a rip off, really a 1000 dollars for a chip? what is it made of pixie dust? Hope AMD gets the market stable on the price, but that also means I'll be waiting at least 2 or 3 years to have a eight-core at a good price, like the rest of us.

Haha, thinking $1000 is expensive.. how cute. A mid-range Xeon is closer to around $2500 nowadays, with the top-end ones cracking $6000

Intel charges what they want because they can, and thanks to mainstream desktop not needing more than 4 cores without HT, there just isn't enough demand for more cores on LGA115x for Intel to move it's slow ass and raise core counts. When they increse the core count again, I have a strong feeling that unlike 1->2 cores or 2->4cores, the performance boost of 4->6 or 4->8 will take a lot longer to materialize for most, if at all.

On the other end of the spectrum, more and more stuff wants a good GPU, and the popularity of phones and tablets demand ever more GPU to smooth animations as far as possible, so the iGPU keeps getting improved.

 

[bpgt64]

Gaming evolved SUCKS.

 

[ZeDestructor]

Gaming evolved SUCKS.

Funny.. I rather like GFE on my end...