AMD Clarifies Why It Uses Intel Core i7 In Its Project Quantum Gaming PC

[InvalidError]

When AMD claims a 40% IPC increase per core, I get suspicious that they're dropping the "module" nomenclature they tried to sell us on.

No reason to be "suspicious" since AMD has confirmed a long time ago that they are switching from their Clustered Multi-Threading modules (multiple somewhat simpler cores sharing some resources) to Simultaneous Multi-Threading (one core simultaneously handling multiple instruction flows) like Intel does.

And yes, I too suspect a fair chunk of that 40% claim comes from SMT packing individual cores' pipelines more steadily. If half of it is raw single-threaded IPC, that would make Zen about even with Haswell. Not ideal but hopefully enough to keep AMD in the game.

 

[siman0]

When AMD claims a 40% IPC increase per core, I get suspicious that they're dropping the "module" nomenclature they tried to sell us on. So, as an eight-core, four-module Bulldozer chip transitions into a quad-core zen chip the IPC doubles like magic.

AMDs current module comprises of 2 cores. They stated it in their product release, its made for multi-threaded applications. Only problem is most programs cant thread span past 4 process threads... That is also why it doesn't have hyper threading. Their CPU design is way ahead of its time, they didn't give a thought to legacy programs or the lack of progression from software developers.

 

[M-31]

Wise business decision AMD. You have nothing 2 be ashamed about.
I'll be taking a serious look at your new 'Quantum Gaming PC'. From what I've read about it so far sounds like you guys have got yourselves a winner.
'CONGRATULATIONS'

 

[somebodyspecial]

Having said that, am I the only one NOT impressed by the size of this supposedly SFF box? Look at it in his hands. It is rather large together and then their external unit has to be figured in too. Doesn't it look about as wide as him and not exactly short to begin with?

I can build my own system and probably take up less space or at least save a chunk of money I'd guess. I though SFF was supposed to take up SMALL SPACE. Why would you want all that wasted space between them and an external unit too? Storing this anywhere makes it a PITA IMHO. Not quite sure why they wasted R&D on this. It wouldn't be a massive money maker no matter how you slice it. Let someone else R&D how to put 2 furys in a box with water. Get back to concentrating on CPU/GPU/DRIVERS period. No more APU (thankfully pulled up ZEN, delayed apu crap that can't make a dime, should've listened to Dirk Meyer ages ago and stayed CORE PRODUCTS), no more crap like this or consoles, just PURE CPU (with high IPC and plenty of die space to hurt Intel with) and pure GPU/drivers.

Well, you could certainly build one, and maybe taking up less space than Quantum. But such aesthetics, and more importantly, all those watercooled parts inside (with maybe an added advantage of OCing them) really puts all those space to use. I mean, it's not too big, not too small. It's fascinatingly a good size to catch eyes, even if it sits somewhere in the corner of a room. Having said that, AMD did make use of its R&D. And as we know of today, chips companies, nvidia and Intel aren't sitting only with CPUs, GPUs, drivers and all. They are making SHEILD, NUCs, and what not. They are expanding their portfolio and it's benefiting them a lot. It's pleasant to see AMD trying to reimagine a new design of living room + gaming SFF, although it isn't quite small, and trying to expand its portfolio. However, it packs all those HPC muscle in a ready-made product, which really impresses me.

True AMD needs to get a good ground on supports, GPU and CPU architecture but it's not going to happen overnight. I presume AMD knows it's losing ground under its feet and taking initiatives, secretively of the crowd, secretively of rivals.

My problem is they're spending on R&D for this type of stuff (cool yeah, but profitable?) instead of having RELEASE drivers for product launches (fury betas?), products actually to give to reviewers for launches (the entire 300 series is a paper launch, nothing even sent to reviewers, card makers did it) etc. AMD is losing money, and can not afford artsy fartsy stuff that steals from CORE PRODUCTS. Fury with 1B more transistors should have release drivers for DAY1 and should have dominated 980ti. How many of these things do you think sell? How many do they have to sell to make a profit on them after covering R&D of such a unique box? I'm sure some people will think it's cool (everyone likes cool looking toys), but how many will pay for it? That is my point. AMD needs to think one thing at this point: Will this product make the company money or not? Because they haven't made a profit in 15yrs (lost 7BILLION!). Over the last 12yrs lost 6B+. At some point you need to FOCUS. Dirk Meyer was right when he said all this junk would steal from CORE PRODUCTS, and they needed to secure top products before branching out to new markets. IE, make some cash before doing fun/pretty toys like this.

NV passed on consoles because it would steal R&D from core products. One look at AMD (no WHQL drivers since Dec 2014, NV has one monthly sometimes 2), and you see this is true. It's a PITY AMD dumped Dirk Meyer, as we would have had ZEN 3+ years ago if he was still there probably (console R&D would have been all CPU/GPU) and APU's wouldn't have been what seems to be their top priority of the last few years. Priority one should have been PURE GPU/CPU and lest I forget, DRIVERS. IE, core products. Nvidia has 3B in the bank and money for shield type stuff, grid etc (and stock dividends, though I hate that! MORE R&D!). Don't forget Grid/shield both cost 10mil each to develop. I'm guessing the rev2 of the handheld (coming shortly) takes less since you can just drop in a 14nm samsung version of x1 with little changes to the device.

It's comic people rate me down for a post about reality (my previous post...LOL). I don't hate AMD, I hate that I can't buy their stock these days due to losing $6B in the last 12 years, and management still doesn't seem to get it yet. Hopefully ZEN is a sign they are going back to CORE products and having more FOCUS. But the Fiji release and whole 300's series seems to say they still are not on track (beta drivers, no reviewers with product, etc). Anandtech still hasn't posted a review of anything...LOL. And they are AMD's shill site, AMD portal and all. I guess they're still looking for benchmarks that can show AMD winning something. Not one peep from them on 390's or fury. That is very telling, when they have a direct line to AMD. They can't get a sample?

Most reviews of anything have shown they had little time to fully test stuff (why toms no ocing etc?), and some have even said this in their articles, like 'due to time constraints, only testing 3 games' etc. Both launches have sucked. Bring back Jerry and Dirk and get some work done 😉 Maybe Su has had more to do with ZEN than I think and this is her new direction (back to Jerry/Dirk type focus), but we'll see. Maybe Fury/300's were not her fault, but the slide showing fury beating 980ti in a dozen games was under her, and they should not have posted that at all! Not when all the reviews say they'd take a 980ti instead and I haven't seen one that showed it winning everything even with 4k which <5% of us use. IT's bad when you give sites a slide and they use it to show how badly you are wrong. Instead of building eye catching stuff that won't make a dime, how about making the SLIDES be TRUE? Then you can charge a premium because you're WINNING. Then you gain back market share due to ...WINNING. I'll take ugly and winning every day of the week over "it looks pretty, but loses".

 

[Gargamel Shmidth]

Not everybody wants AMD FX CPU in theirs Config. So that decision is OK.
FX for AMD fans and i7 for Intel fans -> everybody Happy + You get small powerfull PC for Gaming.
IMO the i7 DC 4cores is not enough for 2x Fiji, the should have put i7 Extreme x6 in that Baby 😀

 

[Frozen Fractal]

Having said that, am I the only one NOT impressed by the size of this supposedly SFF box? Look at it in his hands. It is rather large together and then their external unit has to be figured in too. Doesn't it look about as wide as him and not exactly short to begin with?

I can build my own system and probably take up less space or at least save a chunk of money I'd guess. I though SFF was supposed to take up SMALL SPACE. Why would you want all that wasted space between them and an external unit too? Storing this anywhere makes it a PITA IMHO. Not quite sure why they wasted R&D on this. It wouldn't be a massive money maker no matter how you slice it. Let someone else R&D how to put 2 furys in a box with water. Get back to concentrating on CPU/GPU/DRIVERS period. No more APU (thankfully pulled up ZEN, delayed apu crap that can't make a dime, should've listened to Dirk Meyer ages ago and stayed CORE PRODUCTS), no more crap like this or consoles, just PURE CPU (with high IPC and plenty of die space to hurt Intel with) and pure GPU/drivers.

Well, you could certainly build one, and maybe taking up less space than Quantum. But such aesthetics, and more importantly, all those watercooled parts inside (with maybe an added advantage of OCing them) really puts all those space to use. I mean, it's not too big, not too small. It's fascinatingly a good size to catch eyes, even if it sits somewhere in the corner of a room. Having said that, AMD did make use of its R&D. And as we know of today, chips companies, nvidia and Intel aren't sitting only with CPUs, GPUs, drivers and all. They are making SHEILD, NUCs, and what not. They are expanding their portfolio and it's benefiting them a lot. It's pleasant to see AMD trying to reimagine a new design of living room + gaming SFF, although it isn't quite small, and trying to expand its portfolio. However, it packs all those HPC muscle in a ready-made product, which really impresses me.

True AMD needs to get a good ground on supports, GPU and CPU architecture but it's not going to happen overnight. I presume AMD knows it's losing ground under its feet and taking initiatives, secretively of the crowd, secretively of rivals.

My problem is they're spending on R&D for this type of stuff (cool yeah, but profitable?) instead of having RELEASE drivers for product launches (fury betas?), products actually to give to reviewers for launches (the entire 300 series is a paper launch, nothing even sent to reviewers, card makers did it) etc. AMD is losing money, and can not afford artsy fartsy stuff that steals from CORE PRODUCTS. Fury with 1B more transistors should have release drivers for DAY1 and should have dominated 980ti. How many of these things do you think sell? How many do they have to sell to make a profit on them after covering R&D of such a unique box? I'm sure some people will think it's cool (everyone likes cool looking toys), but how many will pay for it? That is my point. AMD needs to think one thing at this point: Will this product make the company money or not? Because they haven't made a profit in 15yrs (lost 7BILLION!). Over the last 12yrs lost 6B+. At some point you need to FOCUS. Dirk Meyer was right when he said all this junk would steal from CORE PRODUCTS, and they needed to secure top products before branching out to new markets. IE, make some cash before doing fun/pretty toys like this.

NV passed on consoles because it would steal R&D from core products. One look at AMD (no WHQL drivers since Dec 2014, NV has one monthly sometimes 2), and you see this is true. It's a PITY AMD dumped Dirk Meyer, as we would have had ZEN 3+ years ago if he was still there probably (console R&D would have been all CPU/GPU) and APU's wouldn't have been what seems to be their top priority of the last few years. Priority one should have been PURE GPU/CPU and lest I forget, DRIVERS. IE, core products. Nvidia has 3B in the bank and money for shield type stuff, grid etc (and stock dividends, though I hate that! MORE R&D!). Don't forget Grid/shield both cost 10mil each to develop. I'm guessing the rev2 of the handheld (coming shortly) takes less since you can just drop in a 14nm samsung version of x1 with little changes to the device.

It's comic people rate me down for a post about reality (my previous post...LOL). I don't hate AMD, I hate that I can't buy their stock these days due to losing $6B in the last 12 years, and management still doesn't seem to get it yet. Hopefully ZEN is a sign they are going back to CORE products and having more FOCUS. But the Fiji release and whole 300's series seems to say they still are not on track (beta drivers, no reviewers with product, etc). Anandtech still hasn't posted a review of anything...LOL. And they are AMD's shill site, AMD portal and all. I guess they're still looking for benchmarks that can show AMD winning something. Not one peep from them on 390's or fury. That is very telling, when they have a direct line to AMD. They can't get a sample?

Most reviews of anything have shown they had little time to fully test stuff (why toms no ocing etc?), and some have even said this in their articles, like 'due to time constraints, only testing 3 games' etc. Both launches have sucked. Bring back Jerry and Dirk and get some work done 😉 Maybe Su has had more to do with ZEN than I think and this is her new direction (back to Jerry/Dirk type focus), but we'll see. Maybe Fury/300's were not her fault, but the slide showing fury beating 980ti in a dozen games was under her, and they should not have posted that at all! Not when all the reviews say they'd take a 980ti instead and I haven't seen one that showed it winning everything even with 4k which <5% of us use. IT's bad when you give sites a slide and they use it to show how badly you are wrong. Instead of building eye catching stuff that won't make a dime, how about making the SLIDES be TRUE? Then you can charge a premium because you're WINNING. Then you gain back market share due to ...WINNING. I'll take ugly and winning every day of the week over "it looks pretty, but loses".

Well, yeah, I completely agree with you on this. They should be focusing on their core products, R&D and for taking their next step. But I think they are suffering hugely from mismanagement everywhere, and maybe Lisa Su is trying to get everything back on track. Yet it takes time, Su came to CEO not long enough, and Zen seems to hold some TRUE promise for AMD trying to be healed (although I am now quite suspicious whether Zen can hold on to it, after the Fury X incident). It's a big company and changing everything is very, very hard. Maybe they can't even knock those lazy, good-for-nothing people out and getting those posts filled with great minds due to not enough cash on hand, and could be relying on Zen to change all that. If it's not Zen, there's nothing going to pull AMD from this bad fate, imo.

I too feel for AMD, and wish they'd give Intel and nvidia some real brawl soon we enthusiasts are all hungry for.

 

[uglyduckling81]

Having said that, am I the only one NOT impressed by the size of this supposedly SFF box? Look at it in his hands. It is rather large together and then their external unit has to be figured in too. Doesn't it look about as wide as him and not exactly short to begin with?

I can build my own system and probably take up less space or at least save a chunk of money I'd guess. I though SFF was supposed to take up SMALL SPACE. Why would you want all that wasted space between them and an external unit too? Storing this anywhere makes it a PITA IMHO. Not quite sure why they wasted R&D on this. It wouldn't be a massive money maker no matter how you slice it. Let someone else R&D how to put 2 furys in a box with water. Get back to concentrating on CPU/GPU/DRIVERS period. No more APU (thankfully pulled up ZEN, delayed apu crap that can't make a dime, should've listened to Dirk Meyer ages ago and stayed CORE PRODUCTS), no more crap like this or consoles, just PURE CPU (with high IPC and plenty of die space to hurt Intel with) and pure GPU/drivers.

Well, you could certainly build one, and maybe taking up less space than Quantum. But such aesthetics, and more importantly, all those watercooled parts inside (with maybe an added advantage of OCing them) really puts all those space to use. I mean, it's not too big, not too small. It's fascinatingly a good size to catch eyes, even if it sits somewhere in the corner of a room. Having said that, AMD did make use of its R&D. And as we know of today, chips companies, nvidia and Intel aren't sitting only with CPUs, GPUs, drivers and all. They are making SHEILD, NUCs, and what not. They are expanding their portfolio and it's benefiting them a lot. It's pleasant to see AMD trying to reimagine a new design of living room + gaming SFF, although it isn't quite small, and trying to expand its portfolio. However, it packs all those HPC muscle in a ready-made product, which really impresses me.

True AMD needs to get a good ground on supports, GPU and CPU architecture but it's not going to happen overnight. I presume AMD knows it's losing ground under its feet and taking initiatives, secretively of the crowd, secretively of rivals.

My problem is they're spending on R&D for this type of stuff (cool yeah, but profitable?) instead of having RELEASE drivers for product launches (fury betas?), products actually to give to reviewers for launches (the entire 300 series is a paper launch, nothing even sent to reviewers, card makers did it) etc. AMD is losing money, and can not afford artsy fartsy stuff that steals from CORE PRODUCTS. Fury with 1B more transistors should have release drivers for DAY1 and should have dominated 980ti. How many of these things do you think sell? How many do they have to sell to make a profit on them after covering R&D of such a unique box? I'm sure some people will think it's cool (everyone likes cool looking toys), but how many will pay for it? That is my point. AMD needs to think one thing at this point: Will this product make the company money or not? Because they haven't made a profit in 15yrs (lost 7BILLION!). Over the last 12yrs lost 6B+. At some point you need to FOCUS. Dirk Meyer was right when he said all this junk would steal from CORE PRODUCTS, and they needed to secure top products before branching out to new markets. IE, make some cash before doing fun/pretty toys like this.

NV passed on consoles because it would steal R&D from core products. One look at AMD (no WHQL drivers since Dec 2014, NV has one monthly sometimes 2), and you see this is true. It's a PITY AMD dumped Dirk Meyer, as we would have had ZEN 3+ years ago if he was still there probably (console R&D would have been all CPU/GPU) and APU's wouldn't have been what seems to be their top priority of the last few years. Priority one should have been PURE GPU/CPU and lest I forget, DRIVERS. IE, core products. Nvidia has 3B in the bank and money for shield type stuff, grid etc (and stock dividends, though I hate that! MORE R&D!). Don't forget Grid/shield both cost 10mil each to develop. I'm guessing the rev2 of the handheld (coming shortly) takes less since you can just drop in a 14nm samsung version of x1 with little changes to the device.

It's comic people rate me down for a post about reality (my previous post...LOL). I don't hate AMD, I hate that I can't buy their stock these days due to losing $6B in the last 12 years, and management still doesn't seem to get it yet. Hopefully ZEN is a sign they are going back to CORE products and having more FOCUS. But the Fiji release and whole 300's series seems to say they still are not on track (beta drivers, no reviewers with product, etc). Anandtech still hasn't posted a review of anything...LOL. And they are AMD's shill site, AMD portal and all. I guess they're still looking for benchmarks that can show AMD winning something. Not one peep from them on 390's or fury. That is very telling, when they have a direct line to AMD. They can't get a sample?

Most reviews of anything have shown they had little time to fully test stuff (why toms no ocing etc?), and some have even said this in their articles, like 'due to time constraints, only testing 3 games' etc. Both launches have sucked. Bring back Jerry and Dirk and get some work done 😉 Maybe Su has had more to do with ZEN than I think and this is her new direction (back to Jerry/Dirk type focus), but we'll see. Maybe Fury/300's were not her fault, but the slide showing fury beating 980ti in a dozen games was under her, and they should not have posted that at all! Not when all the reviews say they'd take a 980ti instead and I haven't seen one that showed it winning everything even with 4k which <5% of us use. IT's bad when you give sites a slide and they use it to show how badly you are wrong. Instead of building eye catching stuff that won't make a dime, how about making the SLIDES be TRUE? Then you can charge a premium because you're WINNING. Then you gain back market share due to ...WINNING. I'll take ugly and winning every day of the week over "it looks pretty, but loses".

Well, yeah, I completely agree with you on this. They should be focusing on their core products, R&D and for taking their next step. But I think they are suffering hugely from mismanagement everywhere, and maybe Lisa Su is trying to get everything back on track. Yet it takes time, Su came to CEO not long enough, and Zen seems to hold some TRUE promise for AMD trying to be healed (although I am now quite suspicious whether Zen can hold on to it, after the Fury X incident). It's a big company and changing everything is very, very hard. Maybe they can't even knock those lazy, good-for-nothing people out and getting those posts filled with great minds due to not enough cash on hand, and could be relying on Zen to change all that. If it's not Zen, there's nothing going to pull AMD from this bad fate, imo.

I too feel for AMD, and wish they'd give Intel and nvidia some real brawl soon we enthusiasts are all hungry for.

Not releasing a decent high end CPU for 5 years has really hurt AMD.
I'm using an i5 2500k which was better than anything AMD offered in early 2011 and it still is.
It's just crazy the stagnation of the CPU market that has occurred due to AMD not competing at all.

 

[01002920]

wait, what? AMD using a competing CPU?, that makes absolutely no sense as a Marketing Campaign Image guys, I would want to promote the uses and bounderies of my own product rather than promote the competition!. Anyways AMD rules the console market and Intel has a slightly margin over AMD in terms of PC gaming so I think things keep in balance

 

[chris AMD]

I'm currently making due with a fx6300 clocked at 4.5ghz. Looking forward to zen and Am4. I will replace my current setup with that. Still might go Nvidia for graphics card. I want to upgrade from a 270x and nvidia still seems best for a 1080p setup

 

[uglyduckling81]

Anyways AMD rules the console market and Intel has a slightly margin over AMD in terms of PC gaming so I think things keep in balance

A slight margin? In the server world Intel has 97-98% market share. In desktop Intel has approximately 80-85% market share.
AMD aren't even competing. That's classified as a complete monopoly.
The only sector they have any market share is GPU and that's down to 30%.
The console sector is tiny compared to PC. XBone and PS4 have sold about 3-4M units so far this year combined. PC's have sold an estimated 120-150 million units. Consoles are tiny tiny profit margins with low unit sales.

 

[slamm]

Wise maybe, but it's kinda weird. It would be like Ford putting a chevy engine in their new mustang