Wednesday, December 13th 2017

AMD Confirms 2nd Generation Ryzen Processors to Debut in Q1-2018

At a press event, AMD confirmed that its 2nd generation Ryzen desktop processors will debut in Q1-2018 (before April). It also clarified that "2nd Generation" does not equal "Zen2" (a micro-architecture that succeeds "Zen"). 2nd Generation Ryzen processors are based on two silicons, the 12 nm "Pinnacle Ridge," which is a GPU-devoid silicon with up to eight CPU cores; and "Raven Ridge," which is an APU combining up to 4 CPU cores with an iGPU based on the "Vega" graphics architecture. The core CPU micro-architecture is still "Zen." The "Pinnacle Ridge" silicon takes advantage of the optical shrink to 12 nm to increase clock speeds, with minimal impact on power-draw.

AMD is also launching a new generation of chipset, under the AMD 400-series. There's not much known about these chipsets. Hopefully they feature PCIe gen 3.0 general purpose lanes. The second-generation Ryzen processors and APUs will carry the 2000-series model numbering, with clear differentiation between chips with iGPU and those without. Both product lines will work on socket AM4 motherboards, including existing ones based on AMD 300-series chipset (requiring a BIOS update). AMD is reserving "Zen2," the IPC-increasing successor of "Zen" for 2019. The "Mattise" silicon will drive the multi-core CPU product-line, while the "Picasso" silicon will drive the APU line. Both these chips will run on existing AM4 motherboards, as AMD plans to keep AM4 as its mainstream-desktop socket till 2020.
Source: WCCFTech
Add your own comment

101 Comments on AMD Confirms 2nd Generation Ryzen Processors to Debut in Q1-2018

#1
Vayra86
EarthDog said:
Many tests would beg to differ... IPC and clockspeed is important. If IPC wasn't important, we'd all still be running SB at 5 GHz, no?

There are real gains to be had from IPC in both real world testing and benchmarking.

Am I missing what you are saying here? I jumped in the middle of this silly p1ssing contest. :p
I believe the statement was: AMD made a bigger IPC jump, so they have the better CPU. Or more neutral: IPC is a performance indicator.

Neither is true. Its merely one piece of the puzzle and just looking at IPC is a great way to be oblivious about the actual performance.

Best example exists in the Nvidia camp, where people sometimes still say 'Maxwell had better IPC'. Sure, now you try to run your 4K/ultra games on it.
Posted on Reply
#2
notb
Vayra86 said:
IPC only matters for scientific purposes really. It has no bearing on end performance and never has. Its similar to this ridiculous mention of Tflops for GPUs, implying a relation where there is none.
I know that very well.
Why not write an "editorial" on TPU for all the IPC lovers on this forum? :-)
I mean, honestly, other reviewing sites have more "education materials".
EarthDog said:
Many tests would beg to differ... IPC and clockspeed is important. If IPC wasn't important, we'd all still be running SB at 5 GHz, no?
There are real gains to be had from IPC in both real world testing and benchmarking.
CPU engineers look at things like IPC and that's fine, even expected.
But why would users be so interested? It's just a weird indicator. It can be used to compare CPUs, but doesn't say much about actual performance differences.
Posted on Reply
#3
Vayra86
notb said:
I know that very well.
Why not write an "editorial" on TPU for all the IPC lovers on this forum? :)
I mean, honestly, other reviewing sites have more "education materials".

CPU engineers look at things like IPC and that's fine, even expected.
But why would users be so interested? It's just a weird indicator. It can be used to compare CPUs, but doesn't say much about actual performance differences.
I write editorials everyday, they just don't say Editorial on top lol

And yes I Know you know, merely confirming ^^
Posted on Reply
#4
looncraz
Vayra86 said:
IPC only matters for scientific purposes really. It has no bearing on end performance and never has. Its similar to this ridiculous mention of Tflops for GPUs, implying a relation where there is none.
IPC matters A LOT when comparing different generations of products. It allows you to get a general idea of what to expect at a given frequency.

Without it, a 5GHz 2500k/2600k would still be the gamer's choice. Instead we know that Ryzen, Haswell, and Skylake all have higher IPC sufficient to enable better performance at lower clocks.

IPC is just one measure of performance, though, and must be taken into consideration with frequency, core count, and platform (both hardware and software).

When every core from the last half decade (or more), except Ryzen, can reach 4.5GHz easily, IPC becomes the big player.
Posted on Reply
#5
Vayra86
looncraz said:
IPC matters A LOT when comparing different generations of products. It allows you to get a general idea of what to expect at a given frequency.

Without it, a 5GHz 2500k/2600k would still be the gamer's choice. Instead we know that Ryzen, Haswell, and Skylake all have higher IPC sufficient to enable better performance at lower clocks.

IPC is just one measure of performance, though, and must be taken into consideration with frequency, core count, and platform (both hardware and software).

When every core from the last half decade (or more), except Ryzen, can reach 4.5GHz easily, IPC becomes the big player.
No not really, because you forget about core count, you also forget about frequency changes when active core counts go up, and you forget about architectural related software optimization. Same clock CPUs with same core counts from AMD and Intel will still end up in a different order of end performance depending on the application at hand.

Or how about: AVX? Different implementations of Boost/Turbo/XFR you name it. I could keep adding things as I go, but I think the point stands. Even SMT and HT are not identical in how they perform.

And even better: how about looking at Ryzen's internals, and considering the presence of CCX'es and IF that scale heavily off memory bandwidth. There goes your IPC theory... Same clock CPU, same CPU even, but different RAMs...

Do note: 'a general idea of performance' is really saying as much as 'it'll do the job' these days, because as you say, the differences have become a lot smaller gen to gen and even AMD to Intel. For an enthusiast I don't believe 'a general idea' is sufficient, or even remotely satisfying information.
Posted on Reply
#6
EarthDog
Vayra86 said:
I believe the statement was: AMD made a bigger IPC jump, so they have the better CPU. Or more neutral: IPC is a performance indicator.

Neither is true. Its merely one piece of the puzzle and just looking at IPC is a great way to be oblivious about the actual performance.

Best example exists in the Nvidia camp, where people sometimes still say 'Maxwell had better IPC'. Sure, now you try to run your 4K/ultra games on it.
Ahhh, ok, I missed some context. Yeah, that thinking seems a bit off...

IPC is a performance indicator, but not by itself. There is an ecosystem of many things including clockspeed, core count, cache, etc, which paint the big picture.

The fact that AMD made a 50% jump in 4 years isn't much to hang a hat on IMO...
notb said:
CPU engineers look at things like IPC and that's fine, even expected.
But why would users be so interested? It's just a weird indicator. It can be used to compare CPUs, but doesn't say much about actual performance differences.
Because it shows the actual improvement clockspeed and cores notwithstanding. Again, a 50K foot view is needed to paint the bigger picture, but IPC is certainly a consideration... just not the generational gap that loon is hanging his hat on. ;)
Posted on Reply
#7
looncraz
Vayra86 said:
No not really, because you forget about core count, you also forget about frequency changes when active core counts go up, and you forget about architectural related software optimization. Same clock CPUs with same core counts from AMD and Intel will still end up in a different order of end performance depending on the application at hand.

Or how about: AVX? Different implementations of Boost/Turbo/XFR you name it. I could keep adding things as I go, but I think the point stands. Even SMT and HT are not identical in how they perform.

And even better: how about looking at Ryzen's internals, and considering the presence of CCX'es and IF that scale heavily off memory bandwidth. There goes your IPC theory... Same clock CPU, same CPU even, but different RAMs...
As I've said a billion times, there are many different types of performance.

For single threaded performance, it's almost PURELY IPC * Frequency (which includes all turbo implementation details).

Everything else is for multi-threaded performance, which is HIGHLY application dependent - and not owing itself well to generalization... but, generally speaking, more cores and more threads, the better. All other factors are minor (CCX penalty? Like 5%... in the few scenarios where it's visible... Difference in SMT vs HT? 5~10%, AMD SMT usually winning).

IPC * Frequency * Core Count * SMT/HT Scaling * Core/Thread Scaling

This will tell you, in a general sense, of what to expect from different processors in total performance. You ALWAYS need to check out what it does for a specific case. Games, for example, have been receiving targeted improvements by Intel since Haswell. Ryzen lacks these same targeted improvements, so its gaming performance is often worse than the above equation would suggest, but its general performance is right on the mark (when looking at relative systems, not actual performance numbers).

True IPC changes for every application and algorithm within that application, but the above equation will still allow you to approximate what to expect.

A CPU with an IPC of 1.0 and 1GHz will generally be as fast as a CPU with an IPC of 2.0 and 500MHz. Throw in another core, though, for the lower clocked CPU and you have potentially double the performance (not really, due to imperfect scaling, but 85% is GOOD ENOUGH for estimates). Throw in SMT for the first CPU and you are comparing...

IPC 1.0, 1GHz, 40% SMT scaling = 1.4 performance rating
IPC 2.0, 0.5GHz, dual core = 1.85 performance rating

Now you can sort products when evaluating them for your uses without even having those products or reviews for them. You only need to know a few details about how the product works and you only need to estimate ONE or two values using those details. That's how I estimated Ryzen to have > 50% higher IPC than Excavator well over a year before its release (I did this in the forums). I didn't expect Zen to have such high SMT scaling, but that was largely because I didn't expect them to go with the optimal configuration I had in my head (but they did!).

But what do I know? I've only been professionally, and reasonably accurately, estimating hardware performance for a couple decades... but that also means I don't look at things as a normal consumer :p
Posted on Reply
#8
Vayra86
looncraz said:
As I've said a billion times, there are many different types of performance.

For single threaded performance, it's almost PURELY IPC * Frequency (which includes all turbo implementation details).

Everything else is for multi-threaded performance, which is HIGHLY application dependent - and not owing itself well to generalization... but, generally speaking, more cores and more threads, the better. All other factors are minor (CCX penalty? Like 5%... in the few scenarios where it's visible... Difference in SMT vs HT? 5~10%, AMD SMT usually winning).

IPC * Frequency * Core Count * SMT/HT Scaling * Core/Thread Scaling

This will tell you, in a general sense, of what to expect from different processors in total performance. You ALWAYS need to check out what it does for a specific case. Games, for example, have been receiving targeted improvements by Intel since Haswell. Ryzen lacks these same targeted improvements, so its gaming performance is often worse than the above equation would suggest, but its general performance is right on the mark (when looking at relative systems, not actual performance numbers).

True IPC changes for every application and algorithm within that application, but the above equation will still allow you to approximate what to expect.

A CPU with an IPC of 1.0 and 1GHz will generally be as fast as a CPU with an IPC of 2.0 and 500MHz. Throw in another core, though, for the lower clocked CPU and you have potentially double the performance (not really, due to imperfect scaling, but 85% is GOOD ENOUGH for estimates). Throw in SMT for the first CPU and you are comparing...

IPC 1.0, 1GHz, 40% SMT scaling = 1.4 performance rating
IPC 2.0, 0.5GHz, dual core = 1.85 performance rating

Now you can sort products when evaluating them for your uses without even having those products or reviews for them. You only need to know a few details about how the product works and you only need to estimate ONE or two values using those details. That's how I estimated Ryzen to have > 50% higher IPC than Excavator well over a year before its release (I did this in the forums). I didn't expect Zen to have such high SMT scaling, but that was largely because I didn't expect them to go with the optimal configuration I had in my head (but they did!).

But what do I know? I've only been professionally, and reasonably accurately, estimating hardware performance for a couple decades... but that also means I don't look at things as a normal consumer :p
This is what I deem over-analyzing which is a shortcut to knowing exactly nothing substantial about what the product actually does in practice. You've expanded your formula now to take into account all the things I've mentioned, and still you're left with a 'general sense'. Can you see the problem?

How is this relevant? In addition, how many 'purely single threaded' scenarios do you really run in real life? Just about zero. There is always something in the background taking resources.
Posted on Reply
#9
looncraz
Vayra86 said:
I believe the statement was: AMD made a bigger IPC jump, so they have the better CPU. Or more neutral: IPC is a performance indicator.

Neither is true. Its merely one piece of the puzzle and just looking at IPC is a great way to be oblivious about the actual performance.

Best example exists in the Nvidia camp, where people sometimes still say 'Maxwell had better IPC'. Sure, now you try to run your 4K/ultra games on it.
I certainly never made that statement. AMD made a bigger IPC jump, sure, but that doesn't give them the better CPU. Intel is still better, but Intel should have been UNTOUCHABLE by now. They're only 5~7% ahead in IPC (though are comfortably ahead in frequency).

GPU IPC, though, has actually remained unchanged for a LONG time. It's 2.0 for both AMD and nVidia. But GPUs aren't measured by IPC, they're measured by their ability to handle parallel operations (so SP scaling, frequency, bandwidth, workload scheduling, feature support. and so on are the main performance differentiators in graphics).

Vayra86 said:
This is what I deem over-analyzing which is a shortcut to knowing exactly nothing substantial about what the product actually does in practice. You've expanded your formula now to take into account all the things I've mentioned, and still you're left with a 'general sense'. Can you see the problem?

How is this relevant? In addition, how many 'purely single threaded' scenarios do you really run in real life? Just about zero. There is always something in the background taking resources.
Always speak in general terms when speaking of general things. That's science. If we want to look at specific performance scenarios, then we are no longer speaking in generalities, but in specifics... where the same formula STILL holds up, but the IPC and scaling values are derived from the application mix. Everything else is derived from the hardware, so still allows accurate comparison and estimation, as appropriate.

Single threaded scenarios are unavoidable. Nearly every application uses a single thread to control everything - to this very day (blame Microsoft's lack of a forward vision of hardware). Backgrounds task offloading only means we have an even more pure single threaded scenario for any given application, so the formula will bring you even closer to the actual relative performance of multiple products.

Games, for example, only scale so well with extra cores. Once you have four cores, you are pretty much done scaling with every game out there - only a few will show any benefit with more. At that point, you only care about IPC * frequency. Bulldozer/Piledriver had cores aplenty, frequency aplenty, cache aplenty, bandwidth aplenty. But they lacked IPC and core scaling... and we know how well that works out.
Posted on Reply
#10
EarthDog
looncraz said:
Intel is still better, but Intel should have been UNTOUCHABLE by now. They're only 5~7% ahead in IPC (though are comfortably ahead in frequency).
Should... lol... why bother when AMD was stinking up the joint for several years? ;)

I think everyone has made their point here though... over... and over... lol
Posted on Reply
#11
looncraz
EarthDog said:
Should... lol... why bother when AMD was stinking up the joint for several years? ;)
Should you reward a company with your money for the privilege of them milking the same basic thing over and over just because the competition faltered?

Anyone who overclocked a Sandy Bridge to 4.5GHz+ has had no real reason to upgrade until just recently - and that's thanks entirely to AMD and Intel's forced response.

EarthDog said:

I think everyone has made their point here though... over... and over... lol
True.
Posted on Reply
#12
EarthDog
Well, SB has been long in the tooth for a couple of years now, to be honest. And there was plenty of reason from Haswell/Broadwell on up to upgrade (not so much AMD offerings) already. SB holds high end GPUs back several percent in as many titles. Sometimes, none at all of course, but its glass ceiling is real... even at 5 GHz (where it would be less, yes). Is it serviceable? Plenty. But for those looking to maximize their investment, they and past generation AMD CPUs are like wearing cement shoes.

As far as rewarding companies. I don't play games like that. Really, I could care less and surely you would believe I am 'part of the problem' as I have been told so before :). That said, my wallet does the talking and follows the performance. I prefer to pay the premium for a faster (albeit slightly) CPU which overclocks past its own boost. The reality is there is VERY little difference between AMD and Intel CPUs at this time and I am spending where I do not have to, for sure. I get that. But, its a faster CPU. It overclocks more making it several more percent faster than the current iteration of Zen. Zen just pushed things over the edge and I am thankful. However, that doesn't mean I suddenly go for the very close 2nd place in a two man race either. :)
Posted on Reply
#13
Super XP
I'll be looking to upgrade my current gaming PC with this revised node shrink for ZEN. Looking forward to the speed boost for this gen.

looncraz said:
I certainly never made that statement. AMD made a bigger IPC jump, sure, but that doesn't give them the better CPU. Intel is still better, but Intel should have been UNTOUCHABLE by now. They're only 5~7% ahead in IPC (though are comfortably ahead in frequency).
If we simply dip into the past, if it wasn't for AMD pushing Innovation hard (Because they had no choice), Intel would never be where they are today. Intel made huge market errors in the past but had a nice pile of $$$ to help cushion the fall. AMD never had that cushion. This is why AMD was first to many highly innovative techs.
Posted on Reply
#15
notb
R0H1T said:

Just wondering if Ryzen 2xxx itself could feature PCIe 4.0 :ohwell:
Not if they're on a backward compatible AM4. But hey, people praise AMD for keeping the same socket for 4 years. Live with the consequences! :P

Super XP said:

If we simply dip into the past, if it wasn't for AMD pushing Innovation hard (Because they had no choice), Intel would never be where they are today. Intel made huge market errors in the past but had a nice pile of $$$ to help cushion the fall. AMD never had that cushion. This is why AMD was first to many highly innovative techs.
Arguably, Intel is also first to many new ideas. But they're much better in execution.
So when Intel tries something new, it either works and we accept is as a fact, or it doesn't and their bashed.
When AMD tries something new and it doesn't work, everyone is inventing excuses based on low R&D, low cash, small team, unkind media etc.

And another big difference: when Intel copies something from AMD, they usually do it better. When AMD copies something from Intel, they struggle.

BTW: who pushed Intel to innovate in the mobile segment? For the last 5 years, while AMD was designing a great new architecture for another gen of gaming desktops, we had a huge revolution in notebooks and tablets - i no small part thanks to Intel.
Posted on Reply
#16
Gmr_Chick
notb said:
Not if they're on a backward compatible AM4. But hey, people praise AMD for keeping the same socket for 4 years. Live with the consequences! :p


Arguably, Intel is also first to many new ideas. But they're much better in execution.
So when Intel tries something new, it either works and we accept is as a fact, or it doesn't and their bashed.
When AMD tries something new and it doesn't work, everyone is inventing excuses based on low R&D, low cash, small team, unkind media etc.

And another big difference: when Intel copies something from AMD, they usually do it better. When AMD copies something from Intel, they struggle.
Oh boy, is the Fanboy Force strong with this one... I'd hardly call being able to pop new CPUs into the same socket and get them to work with a simple BIOS update a "consequence".

I'm going to play the "AMD isn't nearly as big as Intel" card here, but only because it's kind of true. They aren't as big as Intel, they don't have as much cash -- but only because Intel was HELLBENT on burying them every chance they got by tieing them up in court, bribing OEMs for years, etc, so by the time the FX series came out, AMD was basically on its death bed, much to Intel's delight. By not having AMD around anymore, they could remain stagnant with their CPUs because they would basically have a monopoly over the market -- why improve it when there's no competition, right?

You seem to forget that without AMD, and without Ryzen/Threadripper, Intel would still be popping out dual core i3's, quad core i5's and quad core+HT i7's with no noticeable improvements from the last generation. In some ways, they still are, going from Kaby to Coffee Lake, only now it's with "moar cores" and "moar heat".
Posted on Reply
#17
EarthDog
More cores IS an improvement... this is the amd calling card which forced intel to add more with CFL.

Remember, people dont need more cores and wont for a few more years. A quad with HT is plenty for 95% of users.
Posted on Reply
#18
R0H1T
notb said:
Not if they're on a backward compatible AM4. But hey, people praise AMD for keeping the same socket for 4 years. Live with the consequences! :p


Arguably, Intel is also first to many new ideas. But they're much better in execution.
So when Intel tries something new, it either works and we accept is as a fact, or it doesn't and their bashed.
When AMD tries something new and it doesn't work, everyone is inventing excuses based on low R&D, low cash, small team, unkind media etc.

And another big difference: when Intel copies something from AMD, they usually do it better. When AMD copies something from Intel, they struggle.

BTW: who pushed Intel to innovate in the mobile segment? For the last 5 years, while AMD was designing a great new architecture for another gen of gaming desktops, we had a huge revolution in notebooks and tablets - i no small part thanks to Intel.
Why, is there any particular (technical) reason that the Ryzen SoC can't be updated to PCIe 4.0 whilst keeping the (compatible) chipsets down to PCIe 2.0 or PCIe 3.0, since we already have an example of Ryzen 1xxx & 3xx chipsets?

You can thank Google & ARM for that, also Apple, for Intel sank well over $10 billion dollars in their futile efforts, doing a repeat of their OEM shens via contra revenues.You can also thank phone manufacturers who didn't fall for the hook/line/sinker that HP, Dell et al fell for a decade back.
Posted on Reply
#19
Super XP
notb said:
Arguably, Intel is also first to many new ideas. But they're much better in execution.
So when Intel tries something new, it either works and we accept is as a fact, or it doesn't and their bashed.
When AMD tries something new and it doesn't work, everyone is inventing excuses based on low R&D, low cash, small team, unkind media etc.

And another big difference: when Intel copies something from AMD, they usually do it better. When AMD copies something from Intel, they struggle.
AMD is the Underdog. Has been and probably will be for a very long time. That's why.

I've complimented Intel, just that they would not be where they are today without AMD giving them a nice kick forward. In the end, both needed each other, despite the past shenanigans by Intel, in which AMD was paid by Intel via litigation due to damages.
Posted on Reply
#20
notb
Gmr_Chick said:
Oh boy, is the Fanboy Force strong with this one...
Whatever man. We can call me a fanboy if this makes you happy. Yes, I like Intel. I like their products, I like their image, I like their colour even.
What's wrong with that? Is it bad to be an Intel fanboy - as opposed to being an AMD fanboy? Because I don't see you mocking that part of this community. :-)
I'm going to play the "AMD isn't nearly as big as Intel" card here, but only because it's kind of true.
So?
It doesn't matter if AMD is smaller than Intel. What matters is: are they too small to be on similar tech level?
Is Miele is less innovative than Bosch in home appliances?
Is Porsche technologically behind Volkswagen?

Yes, AMD is smaller than Intel, but they also do a lot less stuff, so it's not that easy to compare. And R&D results are never proportional to firm size. A small company has huge R&D costs. But when it gets larger, it's spending more and more on sales and administration.

Looking at the other competitor: NVIDIA is just 60% larger based on revenue and they also seem to be just in a different league in R&D. Until 2014 AMD was larger.
Actually, here's a fan fact for you: when AMD took over ATI in 2006, ATI and NVIDIA were more or less the same size. If we isolated the GPU part of AMD revenue today (including consoles) it would be maybe 1/3 of the company. So current "ATI" is 5 about times smaller than NVIDIA - after just 10 years under AMD's rule.

Yes, I might sound like an Intel fanboy, but this is not why I criticize AMD. I criticize AMD because - from a business standpoint - it is a failed company. It's like if not much had worked there for the last decade. We all see them struggle, but it's not giving any effects. They can't even keep their chief architects. Raja quit right after Vega launch. Keller jumped off months before Ryzen launch.
And while I see some good points in the Zen architecture, I just can't recommend it based on their promise to support AM4 until 2020. And that's simply because, to be brutally honest, I don't think AMD will make it to 2020. Not in their current form at least.
You seem to forget that without AMD, and without Ryzen/Threadripper, Intel would still be popping out dual core i3's, quad core i5's and quad core+HT i7's with no noticeable improvements from the last generation. In some ways, they still are, going from Kaby to Coffee Lake, only now it's with "moar cores" and "moar heat".
I don't see anything wrong in quad core i5s. I have one myself. And before Zen came along, nor did most people that praise Ryzen today. 4 cores worked. And if you were really in that 1% that needed more, you just got a HEDT platform.
Generally speaking, I believe there are more important issues in the world than how fast gaming desktops are. I'd prefer Intel to put their budget in something else: like storage solutions, AI or IoT. But they had to sacrifice possibly billions of USD to suppress this sudden Zen uprising.
Posted on Reply
#21
theoneandonlymrk
notb said:
Whatever man. We can call me a fanboy if this makes you happy. Yes, I like Intel. I like their products, I like their image, I like their colour even.
What's wrong with that? Is it bad to be an Intel fanboy - as opposed to being an AMD fanboy? Because I don't see you mocking that part of this community. :)


So?
It doesn't matter if AMD is smaller than Intel. What matters is: are they too small to be on similar tech level?
Is Miele is less innovative than Bosch in home appliances?
Is Porsche technologically behind Volkswagen?

Yes, AMD is smaller than Intel, but they also do a lot less stuff, so it's not that easy to compare. And R&D results are never proportional to firm size. A small company has huge R&D costs. But when it gets larger, it's spending more and more on sales and administration.

Looking at the other competitor: NVIDIA is just 60% larger based on revenue and they also seem to be just in a different league in R&D. Until 2014 AMD was larger.
Actually, here's a fan fact for you: when AMD took over ATI in 2006, ATI and NVIDIA were more or less the same size. If we isolated the GPU part of AMD revenue today (including consoles) it would be maybe 1/3 of the company. So current "ATI" is 5 about times smaller than NVIDIA - after just 10 years under AMD's rule.

Yes, I might sound like an Intel fanboy, but this is not why I criticize AMD. I criticize AMD because - from a business standpoint - it is a failed company. It's like if not much had worked there for the last decade. We all see them struggle, but it's not giving any effects. They can't even keep their chief architects. Raja quit right after Vega launch. Keller jumped off months before Ryzen launch.
And while I see some good points in the Zen architecture, I just can't recommend it based on their promise to support AM4 until 2020. And that's simply because, to be brutally honest, I don't think AMD will make it to 2020. Not in their current form at least.


I don't see anything wrong in quad core i5s. I have one myself. And before Zen came along, nor did most people that praise Ryzen today. 4 cores worked. And if you were really in that 1% that needed more, you just got a HEDT platform.
Generally speaking, I believe there are more important issues in the world than how fast gaming desktops are. I'd prefer Intel to put their budget in something else: like storage solutions, AI or IoT. But they had to sacrifice possibly billions of USD to suppress this sudden Zen uprising.
Your bias does matter ,as it marks your comments as biased and therefore tainted.
I could pull up many threads where i said dont get an i5 and thats for the last 3 years, still here on eight fx cores able to play every new release on my value bought pc at 2- 4k, ie good enough, after four or more years actually on all of it working.
Posted on Reply
#22
notb
theoneandonlymrk said:
Your bias does matter ,as it marks your comments as biased and therefore tainted.
That I don't argue with. But I just don't get why I'm being mocked for Intel fanboy-ism, why many here biased towards AMD are pretty much allowed to attack the competition.
So how is it? Is being a supporter of a large, successful corporation a no-no? Or maybe it's OK to support an underdog (this word appears in almost every AMD thread I read)?
Or maybe it's AMD being enthusiast-friendly (as in: you have to tinker for hours to make it work) vs Intel being dull (they just work, they limit overclocking etc)?

I just don't understand the reason why AMD gets so much support on this forum (which they don't deserve), while Intel is bashed all the time and wished to perish? Who's the true underdog here? :-)
I could pull up many threads where i said dont get an i5 and thats for the last 3 years
But why?! What's wrong with those i5s?!
still here on eight fx cores able to play every new release on my value bought pc at 2- 4k, ie good enough, after four or more years actually on all of it working.
We shouldn't really praise PCs for working for 4 years.It should be a standard.
Moreover, 2-4k USD or GBP - I presume? Either way it is a huuuge sum for a PC. It's really no achievement that it works for 4 years. I would expect it to survive 6 years at least (as is true for 2500K-based machines, for example). If you're a frequent gamer going for the newest titles, GPUs age quickly. But a CPU should last for half a decade easily. I expect my i5-7500 (a mid-range quad core) to work pretty well in 2022 - already a while after AMD stops making chips for the long-lasting AM4.
Of course I might change the way I use my main PC and suddenly crave for a 10-core HEDT, but if that doesn't happen, this setup will be fine. I'm pretty sure of it.
Posted on Reply
#23
theoneandonlymrk
notb said:
That I don't argue with. But I just don't get why I'm being mocked for Intel fanboy-ism, why many here biased towards AMD are pretty much allowed to attack the competition.
So how is it? Is being a supporter of a large, successful corporation a no-no? Or maybe it's OK to support an underdog (this word appears in almost every AMD thread I read)?
Or maybe it's AMD being enthusiast-friendly (as in: you have to tinker for hours to make it work) vs Intel being dull (they just work, they limit overclocking etc)?

I just don't understand the reason why AMD gets so much support on this forum (which they don't deserve), while Intel is bashed all the time and wished to perish? Who's the true underdog here? :)

But why?! What's wrong with those i5s?!

We shouldn't really praise PCs for working for 4 years.It should be a standard.
Moreover, 2-4k USD or GBP - I presume? Either way it is a huuuge sum for a PC. It's really no achievement that it works for 4 years. I would expect it to survive 6 years at least (as is true for 2500K-based machines, for example). If you're a frequent gamer going for the newest titles, GPUs age quickly. But a CPU should last for half a decade easily. I expect my i5-7500 (a mid-range quad core) to work pretty well in 2022 - already a while after AMD stops making chips for the long-lasting AM4.
Of course I might change the way I use my main PC and suddenly crave for a 10-core HEDT, but if that doesn't happen, this setup will be fine. I'm pretty sure of it.
2-4k resolution,not cost, my fx cost £159 its value for money is good.
And four cores are already seeing their limits being hit in some games now, by 2022 an i5 will Just be for surfing dude , I'm not saying I5s or any quad are bad just that i wouldn't get one to game on long term now or for the last three years.

And how can you not see that favouring one manufacturer limits the worth of your word.

Im a Manchester united fan do you believe i could comment on a Manchester city match without being too harsh due to bias, not possible, to me their shit regardless,many of my cousin's and family heartily dissagree with me.

It's clear that the sun shines out of intels ass for you but to me a great many issues sully intels name ,but im not biased for the right reasons id happily suggest intel purchase to others and I've two new intel platforms on pre-order.
But the sun Does not shine out of intels ass i assure you.

Oh and four++ years i can't remember purchase date, but flat out folding crunching or mining at 50° that whole time and still no issues is not the same as, games two hours a day or generally off most of the day while at work etc.
Posted on Reply
#24
InVasMani
Now isn't the time to hate on AMD. Right now they are the most competitive they've been since AMD64 on the CPU side. Vega of course could be better, but it's not a total dud at the same time and better than Bulldozer by a mile. Also I really think Ryzen Vega APU will be pretty damn good value in terms of price performance form factor and power draw. It's probably the first APU that truly looks and feels like what most everyone envisioned from the AMD/ATI merger in the first place.
Posted on Reply
#25
Super XP
This is great news. Was going to upgrade my entire setup in a couple months time. This is fantastic. Now I can look forward to this Gen 2 as an upgrade option.
Posted on Reply
Add your own comment