While AMD's Ryzen lineup and pricing has leaked out, only the top three Ryzen 7 processors are available for pre-order (with availability on March 2nd). Starting at $329 for the eight core sixteen thread Ryzen 7 1700, these processors are aimed squarely at enthusiasts craving top-end performance. It seems that enthusiasts looking for cheaper and better price/performance options for budget gaming and work machines will have to wait a bit for Ryzen 5 and Ryzen 3 which will reportedly launch in the second quarter and second half of 2017 respectively. Two six core Ryzen 5 processors will launch somewhere between April and June with the Ryzen 3 quad cores (along with mobile and "Raven Ridge" APU parts) following in the summer to end-of-year timeframe hopefully hitting that back-to-school and holiday shopping launch windows respectively.
Image via reddit (user noiserr). Guru3d has another die shot. Six cores will be created by disabling one core from each CCX.
Thanks to leaks, the two six core Ryzen 5 CPUs are the Ryzen 5 1600X at $259 and Ryzen 5 1500 at $229. The Ryzen 5 1600X is a 95W TDP CPU with six cores and twelve threads at 3.6 GHz base to 4.0 GHz boost with 16MB of L3 cache. AMD is pitting this chip against the Intel Core i5 7600K which is a $240 quad core Kaby Lake part sans Hyper-Threading. Meanwhile, the Ryzen 5 1500 is a 65W processor clocked at 3.2 GHz base and 3.5 GHz boost with 16 MB of L3 cache.
Note that the Ryzen 5 1600X features AMD's XFR (extreme frequency) technology which the Ryzen 5 1500 lacks. Both processors are unlocked and can be overclocked, however.
Interestingly, Antony Leather over at Forbes managed to acquire some information on how AMD is making these six core parts. According to his source, AMD is disabling one core (and its accompanying L2 cache) from each four core Core Complex (CCX). Doing this this way (rather than taking two cores from one CCX) should keep things balanced. It also allows AMD to keep all of the processors 16MB of L3 cache enabled and each of the remaining three cores of each complex will be able to access the L3 cache as normal. Previous rumors had suggested that the CCXes were "indivisible" and six cores were not possible, but it appears that AMD is able to safely disable at least one core of a complex without compromising the whole thing. I doubt we will be seeing any odd number core count CPUs from AMD though (like their old try at selling tri-core parts that later were potentially able to be unlocked). I am glad that AMD was able to create six core parts while leaving the entire L3 cache intact.
What is still not clear is whether these six core Ryzen 5 parts are made by physically disabling the core from the complex or if the cores are simply disabled/locked out in the micro code or BIOS/UEFI. It would be awesome if, in the future when yields are to the point where binning is more for product segmentation than because of actual defects, those six core processors could be unlocked!
The top end Ryzen 7 processors are looking to be great performers and a huge leap over Excavator while at least competing with Intel's latest at multi-threaded performance (I will wait for independent benchmarks for single threaded where even from AMD the benchmark scores are close although these benchmark runs look promising). These parts are relatively expensive though, and the cheaper Ryzen 5 and Ryzen 3 (and Raven Ridge APUs) are where AMD will see the most potential sales due to a much bigger market. I am looking forward to seeing more information on the lower end chips and how they will stack up against Intel and its attempts to shift into high gear with moves like enabling Hyper-Threading on lower end Kaby Lake Pentiums and possibly on new Core i5s (that's still merely a rumor though). Intel certainly seems to be taking notice of Ryzen and the reignited competition in the desktop processor space is very promising for consumers!
Are you holding out for a six core or quad core Ryzen CPU or are you considering a jump to the high-end Ryzen 7s?
I might upgrade to a Ryzen 3
I might upgrade to a Ryzen 3 quad core, to finally get on Sata 3, USB-C and M.2 4 lane bandwagon.
I’m now on an ansient Core 2 Duo E8400 based system with an anemic Qudro fx580/firepro V3900.
Still perfect for precise 2d drafting and drawing, but not 3d modeling and rendering.
$60 Pentium g4560 would also
$60 Pentium g4560 would also be a huge upgrade for that e8400 🙂
Yeah, but with a motherboard
Yeah, but with a motherboard and ram, the total would be almost $100. Why risk it?
It wouldn’t be that much of
It wouldn’t be that much of an upgrade, really. I thought it would be when I did that same upgrade for a customer. Aside from a few Youtube videos playing more smoothly there was nothing a normal user would notice.
Having learned my mistake, my second customer upgrading from a similar system got an AMD quad-core APU. The difference was simply night and day.
Both were using the same SSD with their upgrade (Crucial 275Gb whatever it is – got them on sale cheap, bought four of them :p).
G4560 is the new one with
G4560 is the new one with hyperthreading. Equal to core i3-6100. I’m curious which AMD APU is faster than this today ?
Well it’s not about APUs/SOC
Well it’s not about APUs/SOC today as these first Ryzen SKUs are for the desktop but add a few mor months onto today and you will see Ryzen/Vega APUs and you will weep those delicious tears! You will weep like your bank’s manager has done all these years when you came to empty your bank account to give it to Intel! With that bank transaction leaving you just enough pocket change to afford some mac and cheese to live off of until the next pay day when you still will have to afford more transactions for costly MBs and such.
The 1600X looks great.
The 1600X looks great. Full-size L3 cache and top clock speeds for half the price of the 1800X. All the Rzyen chips look great, but I think the 1600X is the pick of the litter.
Not for me. I do alot of
Not for me. I do alot of content creation + encoding, so 1800X will be a much better buy for me.
For a gamer, 1600X might do it.
Why not wait for the Zen
Why not wait for the Zen workstation variants and the price of a 16 core Zen/workstation SKUs may be very affordable even relative to Intel’s consumer 10 core SKUs. Also For encoding workloads AMD will be introducing for the first time some Interposer based APUs with the Zen cores die wired up via the interposer’s silicon etched traces to a big fat Vega GPU die and HBM2 stacks. If you are creating content then maybe you should wait and price any Zen/Workstation SKUs that will be coming 2H 2017 and early 2018. And workstation Zen CPU only SKUs are going to support at least 4 DIMM channels and the Interposer based APU workstation/server SKUs are going to be a whole new class unto themselves.
I doubt that those server and
I doubt that those server and workstation parts will be cheap. AMD needs to make some money somewhere. They will probably be very competitive with intel’s pricing, but not anywhere near consumer level prices.
Yeah Naples platform will be
Yeah Naples platform will be expensive but its aimed at enterprise that has the corporate money to buy it. AMD will want to charge them as much as possible heh. They need to make miney and suvaidize the consumer zen+ R&D somehow! 🙂
There is bound to be some 12
There is bound to be some 12 or 16 core lower cost Zen workstation variants priced less than the 32 core Naples SKU. I can not see AMD leaving a 16 core workstation variant out of its Pro CPU SKU lineup or even some 12 core Zen Professional SKUs for the low cost server/workstation systems users market. There may be some Pro workstation/server Interposer based Zen/Vega/HBM2 APU SKUs to look at from AMD before the end of the year also.
There will be binned parts depending on if AMD’s 32 core Naples SKUs are are actually made up of 2, 16 core Zen dies joined on a module or if they are making a 32 core monolithic die SKU. But there will be a market for 12 and 16 core Zen workstation variants in addition to the Naples top end SKUs.
Is talk of all Ryzen and not
Is talk of all Ryzen and not just the X models coming with XFR correct?
All models can be OC but only
All models can be OC but only some models have XFR
That’s not what some people
That’s not what some people under NDA are saying, that’s why i asked if all Ryzen and not just the X models come with XFR.
Not cool to ask a site
Not cool to ask a site (PCPer) probably under NDA for possible NDA info. The early leaks showed x chips only had XFR. I guess we will see in a week.
No one is forcing them to
No one is forcing them to answer a question that may or may not be part of an NDA.
Actually the rumor that the
Actually the rumor that the ‘X’ indicated XFR capability was debunked by computerbase.de. It was originally a complete guess by videocardz.com when trying to decipher some of the Ryzen processor names when they were originally confirmed. Ergo, all Ryzen processors have XFR capability! Now we just need to see what impact that will have on them, whether it’s really worth anything given different levels of cooling…
AMD lists XFR on the specs for 1700X and 1800X, but not 1700..
Those that won’t bother to OC
Those that won’t bother to OC manually or care to OC manually probably won’t notice the minor clockspeed increase XFR provides, making it a novelty at best.
Those that manually overclock will vastly exceed XFR’s capabilities, making it useless.
I’m happy they did it, but I’m just not sure that it will really matter to anyone.
I based my article on the
I based my article on the leaks and what AMD has said publicly. Ryan probably knows for sure by now but he cant say. From the leaks I was under the impression The X models had the XFR while the non X models did not. What else would differntiate them? Clocks I guess but the premium between the 1700 and 1700x/1800x seems like there has to be something else there to justify the higher prices. heh. I could be wrong though!
Obviously there could be some
Obviously there could be some confusion but it seems odd people who are going to be selling and reviewing them are saying the same thing, that all three Ryzen come with XFR.
If AMD’s HT works similar to
If AMD’s HT works similar to Intel’s, I’m definitely looking at a Ryzen 3
There is no “HyperThreading”
There is no “HyperThreading” branding on AMD’s CPUs and even Intel’s HyperThreading(TM) is just a branding name for Intel’s version of SMT(Simultaneous MultiThreading). AMD’s Zen/Ryzen(Consumer version of Zen) CPUs support SMT(1). Intel’s version of SMT(under the Marketing name HyperThreading) was derived from the work of Dean Tullsen et al, and it’s best to use the generic computing sciences terminology [SMT] and not any company’s marketing/branding to avoid confusion.
The wikipedia entry:
“While multithreading CPUs have been around since the 1950s, simultaneous multithreading was first researched by IBM in 1968 as part of the ACS-360 project. The first major commercial microprocessor developed with SMT was the Alpha 21464 (EV8). This microprocessor was developed by DEC in coordination with Dean Tullsen of the University of California, San Diego, and Susan Eggers and Henry Levy of the University of Washington. The microprocessor was never released, since the Alpha line of microprocessors was discontinued shortly before HP acquired Compaq which had in turn acquired DEC. Dean Tullsen’s work was also used to develop the Hyper-threading (Hyper-threading technology or HTT) versions of the Intel Pentium 4 microprocessors, such as the “Northwood” and “Prescott”.”(1)
(1)[see entry under Historical implementations]
He never mentioned branding
He never mentioned branding any you clearly knew what he meant. Climb out of your own arse.
You are a marketing monkey!
You are a marketing monkey! How’s that fibbing business going! Obfuscation is marketing’s goal and there are plenty of Numpty Heads out there eating that marketing crap up like total fools.
Don’t use HT for SMT, HT is an acronym for a marketing term/brand HyperThreading(TM). It’s AMD’s version SMT versus Intel’s version of SMT versus IBM’s version of SMT!
FYI, AMD’s SMT has more
FYI, AMD’s SMT has more performance than Intel’s HT
No! AMD’s verson of SMT may
No! AMD’s verson of SMT may have more performance than Intel’s version of SMT but only the benchmarks will tell. The point being please use more computing sciences terminology, Simultaneous Multithreading(SMT), and don’t be one of the Numpy Heads that eat up all that marketing naming/branding Bullcrap. HyperThreading(TM) is just a marketing gimmick done by Intel’s marketing to make folks think that Intel invented something new(SMT) when that’s not really true if you read up on where the SMT technology was actually invented and used.
I have a six year old AMD
I have a six year old AMD Phenom X6 1055T and an RX 480 in this machine, and for the games I play and apps I it runs well enough.
So I would probably be find with an R3 processor. But I think I might treat myself to the R5 1600X.
Though I may go for the R7 1800X if the reviews are good, just to get something that I won’t want to replace for another six or seven years.
I’m really glad I wasn’t
I’m really glad I wasn’t trolled with some horrendous obscene image there. That was cute, thanks.
“Cute”? Are you in delusional
“Cute”? Are you in delusional denial of the harsh truth, perhaps?
Funny because it’s true.
Funny because it’s true. After the NDA lifts, shizz is gonna go crazy.
Say HIS name. SAY IT.
Say HIS name. SAY IT.
Now for the Zen/Vega APUs
Now for the Zen/Vega APUs with the Vega NCUs. How many NCUs will AMD be able to get at 14nm relative to the amount of CUs(8) that AMD got on its previous generation APUs at 28nm. The x86 Zen core takes up even less space than an Intel x86 core in the mm^2 metric so AMD should be able to get at least 12 Vega NCUs in addition to 4 Zen cores. The really big question regarding any consumer APU SKUs that use Zen/Vega will be if AMD decides to include some HBM2 to supply the Integrated Vega graphics with plenty of effective bandwidth and also how will the Vega high bandwidth cache controller IP help with APU graphics workloads.
16 on the top end APU would
16 on the top end APU would be awesome :). 12 is much more likely to actually happen though.
I’m going for the 1600X,
I’m going for the 1600X, since any media creation I do will be infrequent, compared to the amount of gaming I do.
That said, I was hoping to get a Ryzen build before Mass Effect: Andromeda came out, but I guess I’ll be able to get a pretty good example of the performance boost a 6c/12t CPU could provide a Frostbite game.
Upgrading to only play the
Upgrading to only play the latest POS from bioware? lmao
Just when you thought you
Just when you thought you were all f_pped out, there’s more.
I hear that xtenz can help
I hear that xtenz can help with recovery time.
For now I can’t see a reason
For now I can’t see a reason to upgrade from my 8370 at this point, but this ryzen is VERY good news for everyone, even putting it’s competitor on their toes.
You’ve got to be kidding me.
You’ve got to be kidding me. One mid-tier “gen1” RyZen CPU is, like, AT LEAST 3 of those “FX 83xx” crap-outs. There’s absolutely NO “good enough” or any logical reason whatsoever at all to stay on that old, hot, power-hungry, underpeforming GARBAGE! What did you pay for that POS, like, 250$, when it came out? You can get 1600X for that. And it’ll blow the flying Fs of that “pseudo-more cores” FX 8370 trash.
The 8370 wasn’t a bad
The 8370 wasn’t a bad processor as far as 28 nm processors are concerned. It just isn’t going to compete with 22, 20, or 14 nm parts though.
Anything that’s made of
Anything that’s made of “pseudo-more cores” half-assed FPUALU corncob MHz “package”, is GARBAGE.
I still have an 8150 in one
I still have an 8150 in one of my backup rigs, and while it’s not the fastest chip in the world, it certainly is still a moderately capable chip.
When my buddies come over for LANs we use that, along with an even older rig with a Phenom 955, and they both still manage decent frame rates in modern games.
My facepalm is the
My facepalm is the size of the planet Jupiter right now
Why not? 32nm Sandy Bridge
Why not? 32nm Sandy Bridge still competes with CPUs on smaller process nodes.
When will people realize that feature size is mostly marketing bullshit?
First, I didn’t pay for it,
First, I didn’t pay for it, second I wasn’t saying that ryzen wasn’t a MASSIVE upgrade on what I have now, I was saying that I have all I need right now, I wont be upgrading for a while.
>I didn’t pay for
>I didn’t pay for it
>I won’t be upgrading for a while
I understand your opinion if
I understand your opinion if the 8370 provides all the performance that you currently need, but it’s not even a contest – whether single or multi-threaded, any Ryzen CPU will destroy the 8370, burn its corpse, and drink the tea made from its ashes.
The most ironical thing about
The most ironical thing about this, is that I remember reading or hearing some rumors flying around not-so-long ago that AM4 is supposedly backwards compatible with AM3+ stones AT THE VERY LEAST. If this is true, this means our here “FecalX 8370 is enough in 2017”-clueless m8 could easily just buy a decently-priced “gen1” Zen motherboard (probably A320 of B350, considering what a cheapskate that d00d is), put that POS of his in there and then relatively painlessly upgrade to any RyZen a little bit later.
You can use AM3+ coolers with
You can use AM3+ coolers with an adapter but not sure how they would make the AM3+ CPUs work in the 1331 pin AM4 socket?
No, can’t use am3+ CPUs in
No, can’t use am3+ CPUs in AM4 motherboards.
2nd, there is no 3x!!! the perf over FX-8000-series. It’s 52% in singlethreaded perf. Multi-threaded perf is pretty close to Intels, considering AMD’s 8000-series is from 2012.
Whats your FPS in GR:Wildlands benchmark in Very HQ? With Temporal AA turned off. Mine is 50 fps avg with a 290x and FXemail@example.com Ghz. Pretty much inline with what everyone else on the Intel side is getting with similar hardware.
Lisa clearly said “52% and
Lisa clearly said “52% and even more”. 52 is the minimal spec, basically. They’re probably holding out better binned stones that exceed the “52%” for either “9” HEDT or “Zen+”.
Actually I thought they
Actually I thought they claimed during that live stream with Lisa Su recently that the 52% was compared to Excavator? Excavator cores were a decent bit faster than Piledriver. So versus the fx 8000 vishera chips it is probably more than a 52% increase though I am not sure how much more. Possibly not 3x though..
Anonymous is right. There
Anonymous is right. There was a slide in the presentation giving full details of the 52% figure.
AMD benchmarked Ryzen vs. both Piledriver and Excavator on SpecInt 2006 using GCC 4.6 and Ubuntu 16.x (they didn’t specify .04 or .10), using internal motherboards and with all CPUs locked to 3GHz.
Ryzen scored 52% better than Piledriver and 64% better than Excavator on single-threaded tests. That does seem like it’s the wrong way ’round, so maybe the difference was due to cache (none of the shipping Excavator parts have L3 cache).
On the 3x figure, though, that’s not as unreasonable as it sounds. Bulldozer and Piledriver shared a decode unit per module, and it really throttled multi-threaded performance, down to around 1.25x rather than 2x in my own tests (running business workloads on Opteron servers). Which is about what you get from SMT.
So with 52% better IPC and twice the threads, you could get close to 3x the multi-threaded perf – depending on your workload.
Thank you, Captain
Thank you, Captain O.
Everyone now knows that Faildozer is a horrendous POS garbage, tanks as always.
And Vishera was just a tiny increment off of that.
Yeah, my point was pretty much coming off of that.
Ah okay, thanks for clearing
Ah okay, thanks for clearing that up! The absence of the L3 cache could be why Zen is 64% better than Excavator but only 52% better than Vishera. Zen is definitely going to blow away Bulldozer. I wonder where we would be had AMD not bet the farm on CMT and instead had continued down the path of the Phenom II architectures? In hindsight Intel definitely had the better design and plan with Core era arches while AMD struggled to get a good implementation (bulldozer didn't hit the clocks they wanted and ran hot) and developer adoption. OTOH if Excavator cores had been the first CMT/xx-dozer parts to come out it might have gone somewhere :-).
I hope that you DO know that
I hope that you DO know that Phenom III actually existed (but never left the lab, because they hit the wall due to technological limitations of that time period, just like Intel with NetBUUUUURRRRRRRRst back in the days)?
Zen is essentially “Phenom IV”. A return to the core (pun intended) principle, but with adjustment to the modern technologies. Phenom III was doomed from the start, that’s why AMD’s BoD lost their goddamn minds and pressured their R&D to work on that FPUALU-corncob POS for so many years. AMD’s BoD was literal “EXECUTIVE CANCER” through the entirety of the “FX” platform’s existence. I was saying it many times before, and will never stop repeating it over and over again: the ONLY major difference between AMD and Intel in the late 2000s and early 2010s was purely only in that Intel already stepped on the rakes that were NetBUUUUURRRRRRRRst/Prescott/”MUH LONG(and hot)PIPE”, Intel learned from their mistakes and completely reworked everything essentially rising like a phoenix from ashes by introducing Core2, while AMD, on the other hand, had ALL of that “very bad, but very educational at the same time” experience still lying ahead of it, AMD was yet to experience the humiliation similar to what Intel felt back in the days of effing Pentium 4. The only difference between them was simply in that Intel was the first one with that, nothing less or more than just that. And then Phenom III “didn’t fly”, AMD’s BoD gone crazy and FX was “born”. A horrible abomination of technological world. This is where AMD finally stepped on rakes too, just like Intel did a few years before them. I’m NEVER joking when I’m saying this: the entire FX platform is LITERALLY “AMD’s NetBUUUUURRRRRRRRst/Pentium 4/Prescott”. Always was from the very beginning, and always will be remembered as being such at the very least by me personally. Intel was caught in that trap in the early and mid-2000s, they’ve had to completely rework everything to get out of it…but AMD’s time to fall in it came much later, with FX. And, just like Intel back then, AMD had to completely rework everything to rise again. And so, Zen was born. Praise be our one and only lord and savior, the everlasting and almighty, Killer Jim. CSW! C-S-W!! It’s the “K6~K8” and Phenom II times all over again, folks. AMD is fully back into the action. Zen IS Phenom IV.
Are you doing that on purpose?
I’ve been itching for a
I’ve been itching for a reasonably priced 8 core processor for years. It’ll make an epic VMware host.
why are there so many
why are there so many options. Same with GPU’s. isn’t it really just 4 pricing tiers
Maybe it should be $50 increments but really just skip the starbucks for a week and $50 increments are pointless.
Because there are enough dies
Because there are enough dies with defective cores/units! So Binning the defective dies will make for some lower cost options like 6 core Zen parts and such instead of completely wasting an entire die. There are even some 8 core Zen parts with no defects that can be clocked higher than other 8 core Zen parts that also have no defects and the 8 core Zen dies that can be clocked the highest become the top binned parts.
Every CPU/GPU/Other processor die undergoes a binning process because even the chip’s difussion process has variances. So some chips come out of the oven with a little better performace characteristics than other cores from other batches or even different difussion lines.
I say bin to the max, why let a processor die go to waste because of 1 or 2 bad cores! Or even 4 bad cores on an 8 core chip is still good for some lower cost 4 core options!
Bin them puppies good!
Wait later in the year. At
Wait later in the year. At the most recent conference it was stated that on March 2nd we’d get Ryzen’s FULL line-up, when did that change? Since when is the R5 line not coming until April-June???
Probably saving up the
Probably saving up the harvested 8 core dies so they can do a huge R5 launch. $329 might be cheap to some people, but I’ll bet they’ll sell 2-3x more R5s, easily.
1700X for me, I previously
1700X for me, I previously owned the 1055T/8120/8320/8350, cannot wait until next week. I am not a big fan of the 4790k I currently run in my build, really really looking forward this release.
R7 1700 is cheap enough
R7 1700 is cheap enough considering the prices of 6700K in the past and 7700K today. So, whoever was thinking for a 7700K shouldn’t worried that much about paying $329 for a 1700. Just wait a few days and see final reviews.
If the chips binned as 1700s
If the chips binned as 1700s have good manual overclocking headroom they are going to be the chip to get!
Ryzen, sounds great, but I am
Ryzen, sounds great, but I am holding out for the modest improvements x299 is going to offer. ” drops mic and runs away from fan boys”