[EDIT]hmmm.....whats going on at AMD? (and history of 3Dfx Rampage)

[Mousemonkey]

I wonder how being able to OC Intel's non K Skylake CPU's is going to affect Zen?

I think the larger question is:

Now that the world has pretty much given up on AMD processors, would a Zen chip that is only equal or marginally better than intel lure enough people back?

So many years of fail might have AMD out of the game before they even start playing.

BTW- welcome Mousemonkey, good to see you here!

Thanks mate and it's good to be back, I had to re-register but that's understandable all things considered. As for Zen the future looks a bit on the bleak side if you ask me, there are a couple of benchies floating round that show the i3 6100 CPU keeping up with and sometimes getting ahead of the FX 8320E.

http://www.techspot.com/review/1108-inte...rclocking/

Of course Intel could shit a brick and put a stop to this as it may impact the sale of their K series CPU's but then why would they? And how? It's the motherboard manufactures that have come up with this little humdinger after all, completely independent of any Intel influence.

Wow!!

I never thought we would see this day!

Has the bios been released to the public yet?  I haven't been keeping up, crazy busy these past few months

No idea mate but I'm trying to find out for you.

[Mousemonkey]

I have asked one of our staff writers to reach out to motherboard manufactures in order to find out who is going to release the BIOS updates necessary and will keep you posted on anything I find out.

[SteelCrysis]

This is AMD's response to the Skylake i3s: http://www.techpowerup.com/218349/amd-in...essor.html

[SteelCrysis]

http://www.extremetech.com/gaming/219798...gainst-amd
Cooler Master says the Fury X cooler was found to not infringe.

Edit: Also this: http://www.techpowerup.com/218578/samsun...-node.html

[SteelCrysis]

http://techreport.com/news/29480/amd-del...er-of-2016
Dual GPU Fiji card is delayed to match the release of VR headsets.

[SteelCrysis]

http://www.techpowerup.com/218898/amd-de...cture.html
AMD's next generation of GPUs is codenamed Polaris, due mid-2016.

http://www.techpowerup.com/218897/micros...d-soc.html
Polaris could be used in an Xbox One refresh.

http://www.tomshardware.com/news/amd-pol...30823.html
More info about Polaris, it's going to use FinFET.

[SickBeast]

Very interesting, thanks steelcrysis.

Sent from my XT1032 using Tapatalk

[SteelCrysis]

Global Foundries will be making Polaris: http://www.extremetech.com/gaming/220603...laris-gpus

[SteelCrysis]

AMD standardizing on one socket: http://www.tomshardware.com/news/single-...30949.html

[Mousemonkey]

http://www.tomshardware.com/news/overclo...30962.html

[SickBeast]

For what reason?

[SickBeast]

The links aren't working so I had to google it and all I could find was that they are standardizing the socket which sounds very good to me.

[SteelCrysis]

R9 Nano price cut down to USD 499: http://techreport.com/news/29584/amd-sla...tag-to-499

[ocre]

Disables C states and hyperthreading amongst  other things.  Did you not read the linked article?  Some processor instructions won't work either (article doesn't specify but would not be surprised to learn it affects AVX).

When a full blown K series i7 is cheaper than flagship quad core (not hex core) Phemon II's were and Z series motherboards are over a third less pricey than good nForce boards used to be I totally and utterly fail to see the point, unless you are a fan of wasting power for no good reason.

That article was full of errors.... Actually, it is complete and utter trash.
Shame on them for being so freaking sloppy, if that is the even the culprit. The fact that it is still up and still says the same words, even though they have been corrected on this, it is strangely suspicious.

Asrock set to correct Toms and what a massive blunder. I do not understand how the older Toms article still exist, it is so wrong, complete misinformation. Here is the update after asrock set to correct Toms:

http://www.tomshardware.com/news/asrock-...31000.html

The significant thing to remember here is that the original Toms article only listed two manufacturers that would have overclock abilities for nonZ170 boards. Only asrock and sun micro. And out of those, they claimed asrock boards would disable hyperthreading. That is completely incorrect. Bogus.

Not only that, when you use asrock boards and don't overclock, nothing is disabled. It's just like a normal board. When you do overclock, you loose insignificant features. C states, which overclockers typically turn off anyway, is not some grand compromise. Turbo? Overclocking usually is done with turbo off.

There is very little loss. The boards run like normal boards when not overclocked. Tom should be embarrassed by the ignorance and should be sued for the massive disservice they have done with their original rubbish

[SteelCrysis]

That article was full of errors.... Actually, it is complete and utter trash.  
Shame on them for being so freaking sloppy, if that is the even the culprit.  The fact that it is still up and still says the same words, even though they have been corrected on this, it is strangely suspicious.

Asrock set to correct Toms and what a massive blunder.  I do not understand how the older Toms article still exist, it is so wrong, complete misinformation.  Here is the update after asrock set to correct Toms:

http://www.tomshardware.com/news/asrock-...31000.html

The significant thing to remember here is that the original Toms article only listed two manufacturers that would have overclock abilities for nonZ170 boards.  Only asrock and sun micro.  And out of those, they claimed asrock boards would disable hyperthreading.  That is completely incorrect.  Bogus.

Not only that, when you use asrock boards and don't overclock, nothing is disabled. It's just like a normal board.  When you do overclock, you loose insignificant features.  C states, which overclockers typically turn off anyway, is not some grand compromise.  Turbo? Overclocking usually is done with turbo off.

There is very little loss.  The boards run like normal boards when not overclocked.  Tom should be embarrassed by the ignorance and should be sued for the massive disservice they have done with their original rubbish

[SteelCrysis]

Didn't mention this earlier, but AMD is introducing a new stock CPU cooler, but according to HardwareCanucks it won't completely replace the old garbage AMD stock cooler: http://www.tomshardware.com/news/amd-wra...31015.html

[SteelCrysis]

First 2 Polaris cards are named Polaris 10 and Polaris 11, and they support HDMI 2.0: http://www.guru3d.com/news-story/amd-wor...is-11.html

[SteelCrysis]

And AMD loses money yet again: http://www.techpowerup.com/219376/amd-re...sults.html

[SteelCrysis]

And now ASRock has decided to disable overclocking on non-Z170 motherboards: http://www.tomshardware.com/news/asrock-...30915.html

[SickBeast]

I really hope Zen is good. All this disabling of overclocking is nonsense. I'm sick of Intel's arrogance. I'm really pulling for AMD. I don't think they can pull it off but we will see. I'm expecting Sandy Bridge or maybe Ivy Bridge IPC performance. Even that is pushing it though. That being said, I think a lot of people could live with Ivy Bridge, particularly if it can clock really high (which it should on the new node).

[SickBeast]

Yes but they also control which chipsets can overclock. These are all artificial limits. Notice how they have disabled it on 95% of the cheaper hardware. The g3258 was an anomaly and it is crippled to boot.

[SickBeast]

Changing the multiplier has nothing to do with the timings. People who know what they are doing can overclock a CPU quite nicely on non-Z boards that are unlocked by the manufacturer. I had a board like that myself, it was awesome. Locking down the boards is done out of sheer green on behalf of Intel.

I realize that you hate AMD but you need to stop showing Intel so much blind love. They have been resting on their laurels since Sandy Bridge was released and at the same time they have been locking down overclocking and basically forcing enthusiasts to pay $200+ for a CPU. Yes, AMD sucks, but Intel has become arrogant and we need some competition to bring the market back in balance. I know I find it alarming personally to see how close smartphone and tablet CPUs are getting to what Intel makes. We are witnessing the erosion of our hobbies. If things keep going like this, desktop PCs are going to be completely dead pretty soon.

[SteelCrysis]

Dual Fiji card spotted: http://wccftech.com/amds-dual-fiji-based...tted-vrla/

[SickBeast]

Ugh. I hate cards like that.

[BoFox]

Rumors had it coming out in Q4 2015 - December at the latest, but AMD missed the Xmas season once again.

I really hope AMD gets it all back together on the 14nm GloFo process, but the bird tells me that Zen isn't quite fitting to Intel's Yang.  It needs to be Yin.

[SickBeast]

Yawn. Please wake me up when Polaris and Pascal arrive. It's going to be a long and boring time until then.

[SteelCrysis]

AMD releases yet more CPU refreshes, Wraith cooler will come with the FX-8370, 2 new CPU coolers from AMD: http://www.tomshardware.com/news/amd-ath...31112.html

Edit: TechReport says one of the CPUs is an Excavator CPU: http://techreport.com/news/29669/amd-put...lon-x4-845

[SteelCrysis]

More info on Excavator: http://www.anandtech.com/show/10009/amd-...845-for-70

[BoFox]

More bulldozer derivates this close to Zen releasing?!  Seriously AMD, what are you thinking?!  Are you even thinking at all?!

Zen isn't due until the end of 2016 at the earliest.  AMD is just real slow with clearing the inventory (trying to get rid of all Piledriver crap first, and then Steamroller..) - plus they didn't make that many Excavator quite yet due to low anticipated demand or perhaps they delayed it a bit to double the L2 cache for the desktop version??
Some are saying that it's supposed to have only 1MB cache per module like the mobile-version Excavators, rather than 2MB, for a total of 4MB - that it's a blatant lie by AMD. 

Maybe AMD wanted to save the thunder for Zen, hoping it would be ready much earlier in 2016, but then realized that with such limited budget and resources, they wouldn't be able to get Zen ready until later.

Here's probably the only serious article about comparing the IPC improvements of Excavator vs Steamroller vs Piledriver (all at the same clock, multi-core and with 1 core):
http://www.planet3dnow.de/cms/18564-amd-...cinebench/

(the link has many more pages of benches)
Not too shabby - the improvements are quite large, eh?  The AVX2 improvements are nice, but it probably throttled in one instance with the 15W TDP ceiling (if that's what others were saying)?

This is how Bulldozer should have been performing in the first place, if AMD had Intel's R&D budget.  It's just not so exciting for AMD to say that they finally fixed it on 28nm after so many years...

[BoFox]

Also, what's so exciting if AMD can't ever hope to touch the unlocked Pentium G3528 for $70 (if a gamer wishes to avoid the minimum frame rate dips that put their watercooled 5GHz FX's to shame compared to the first gen LGA1366 Core i7..):

http://www.newegg.com/Product/Product.as...6819117374

It's really  more geared towards budget-oriented users who don't care about playing games, concerning the i3 competition.

The only thing AMD's got going for it is the Zen hype, after all.

[SteelCrysis]

AMD Wraith cooler doesn't quite live up to AMD's hype about quietness, but rivals Thermalright Macho Rev.B in cooling performance: http://www.tomshardware.com/reviews/amd-...450-2.html

[SteelCrysis]

More info on Polaris: http://wccftech.com/amd-polaris-architec...imum-spec/

[BoFox]

Looks like it will be like the next HD 3870 (55nm) that AMD thought was going to rock the world vs Nvidia's 65nm 8800GT, as AMD released a dual-GPU 3870x2 about a month after the paper launch of a single-chip 3870, just in time for Xmas 2007.  Still, no dice at all as 8800GT took the spotlight, while NV didn't release a dual-chip 9800GX2 until many months later. 

Then AMD finally got the bigger 4870 ready on 55nm with first GDDR5 (which should've been called QDR instead of DDR) just in time for Nvidia's monolithic 65nm GTX 280.  The 4870 actually turned out to be a big success with the humble pricing - probably not something AMD would ever do again if greed continues to run rampant.  

AMD's compact 14nm Polaris vs Nvidia's 16nm Pascal...  history lesson: it doesn't matter which is smaller - what matters is price and performance.

[SteelCrysis]

Possible Fury X2 sighting: http://videocardz.com/58343/is-this-the-...r9-fury-x2

[SteelCrysis]

Polaris will be released in the middle of this year, Zen will be released later this year: http://www.tomshardware.com/news/amd-zen...31338.html

[RolloTheGreat]

Polaris will be released in the middle of this year, Zen will be released later this year: http://www.tomshardware.com/news/amd-zen...31338.html

If Zen surpasses my 4790K by a decent margin, I'd do an AMD build for my next box.

Same with Polaris, I'm a buyer if it beats Pascal.

This is AMDs last stand is my guess.

[BoFox]

Polaris will be released in the middle of this year, Zen will be released later this year: http://www.tomshardware.com/news/amd-zen...31338.html

If Zen surpasses my 4790K by a decent margin, I'd do an AMD build for my next box.

Same with Polaris, I'm a buyer if it beats Pascal.

This is AMDs last stand is my guess.

I'd love to go with Zen if it beats Haswell, and is cheaper than what Intel could very much do with Skylake's enhanced L4 cache by rolling out a part with 128MB eDRAM:

We have more reasons to be excited over the eDRAM in Skylake than what we saw before in Haswell with the i7-4950HQ on mobile and Broadwell on desktop with the i7-5775C, i5-5765C and the relevant Xeons. With the older platforms, the eDRAM was not a proper bidirectional cache per se.  It was used as a victim cache, such that data that was spurned from the L3 cache on the CPU ended up in eDRAM, but the CPU could not place data from the DRAM into the eDRAM without using it first (prefetch prediction). This also meant that the eDRAM was invisible to any other devices on the system, and without specific hooks couldn’t be used by most software or peripherals.

http://www.anandtech.com/show/9990/skyla...-with-gt4e

With Skylake, this changes, the eDRAM lies beyond the L3 and the System Agent as a pathway to DRAM, meaning that any data that wants DRAM space will go through the eDRAM in search for it. Rather than acting as a pseudo-L4 cache, the eDRAM becomes a DRAM buffer and automatically transparent to any software (CPU or IGP) that requires DRAM access. As a result, other hardware that communicates through the system agent (such as PCIe devices or data from the chipset) and requires information in DRAM does not need to navigate through the L3 cache on the processor.  Technically graphics workloads still need to circle around the system agent, perhaps drawing a little more power, but GPU drivers need not worry about the size of the eDRAM when it becomes buffer-esque and is accessed before the memory controller is adjusted into a higher power read request. The underlying message is that the eDRAM is now observed by all DRAM accesses, allowing it to be fully coherent and no need for it to be flushed to maintain that coherence. Also, for display engine tasks, it can bypass the L3 when required in a standard DRAM access scenario. While the purpose of the eDRAM is to be as seamless as possible, Intel is allowing some level on control at the driver level allowing textures larger than the L3 to reside only in eDRAM in order to prevent overwriting the data contained in the L3 and having to recache it for other workloads.

Any benchmarks so far on the Skylake Xeon with Iris Pro?  I'm gonna try to find some..



As to Polaris vs Pascal..  I think I'll just go with Pascal, even if Polaris is 10% faster and 10% cheaper.  I just like Nvidia better these days (PhysX, Gsync, better driver support for almost everything, etc..).

[ocre]

The rabid AMD fans ruin it for me.  It's like support for eating toxic waste.  It is hard to stomach.  

But, I guess I will consider AMD....if they do surpass in performance.

Bofox, I hear you talking about skylake xeon, but what about broadwell e?

[SteelCrysis]

ASRock is not giving up on Intel base clock overclocking: http://techreport.com/news/29823/asrock-...erclocking

[SickBeast]

ASRock is not giving up on Intel base clock overclocking: http://techreport.com/news/29823/asrock-...erclocking

Wow that's awesome! If it works that could be a huge deal! What's the cheapest locked quad core? Even i3s would be overclockable. Very cool.