Can AMD salvage QFX with an in-house chipset?

Page 10 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
I dont agree with BM because I dont think AMD has to salvage QFX for any chipset. Lets look at what AMD set out to do with 4X4. Beat C2D and even the lowest FX-70 does just that at a lower current price than the top C2D. Did 4X4 loss against the C2Q? Its a mobo so it really must be judged for what it was designed. It offers about the same GHz per GHz increase as C2Q with C2Q getting the best percentage on games and the 4X4 on multitasking.

What are the downfalls of both implementation of quad core? The 4X4 increases memory latency and costs more on the motherboard side. The C2Q6700 losses about 600MHz off its top OC over C2D's 4GHz and about 370MHz off its top released C2D6800.

What are the gains for both implementations of quad core? The FX-74 gains in refinement to give an added 200MHz over AMD's top FX-62. The 4X4 has a better upgrade path as its set for 2X-Agena FX's. The C2Q motherboard is cheaper than the 4X4 and one isnt forced to upgrade if Agena FX's performance is worse than expected.

Which company losses here? Both companies loss because this is a tech war. Intel losses due to having to modify a core well short in the cores life cycle and its top C2D6800 falling to 6th over all in about 4 months. AMD losses because the 4X4 and FX's will cut into its opty sales and could be a bad return on investment in the short term.

Which company wins? Neither because only we the consumer have won with cheaper faster CPU's.
 
I dont agree with BM because I dont think AMD has to salvage QFX for any chipset. Lets look at what AMD set out to do with 4X4. Beat C2D and even the lowest FX-70 does just that at a lower current price than the top C2D. Did 4X4 loss against the C2Q? Its a mobo so it really must be judged for what it was designed. It offers about the same GHz per GHz increase as C2Q with C2Q getting the best percentage on games and the 4X4 on multitasking.

What are the downfalls of both implementation of quad core? The 4X4 increases memory latency and costs more on the motherboard side. The C2Q6700 losses about 600MHz off its top OC over C2D's 4GHz and about 370MHz off its top released C2D6800.

What are the gains for both implementations of quad core? The FX-74 gains in refinement to give an added 200MHz over AMD's top FX-62. The 4X4 has a better upgrade path as its set for 2X-Agena FX's. The C2Q motherboard is cheaper than the 4X4 and one isnt forced to upgrade if Agena FX's performance is worse than expected.

Which company losses here? Both companies loss because this is a tech war. Intel losses due to having to modify a core well short in the cores life cycle and its top C2D6800 falling to 6th over all in about 4 months. AMD losses because the 4X4 and FX's will cut into its opty sales and could be a bad return on investment in the short term.

Which company wins? Neither because only we the consumer have won with cheaper faster CPU's.


I only said salvage because if I had said improve there would have been a different discussion. I don't think it needs any help. People who want lower power will NOT GET DX10 SLI.

People who depend more on the CPUs won't get high end single GPUs. Voila, power is down.

It's a SERIOUSLY POWERFUL platform and tweaking will only make it better. As DDR CPUs disappear, DDR2 memory will drop in price, so by the time Vista hits higher clocks will be accessible to most people.
 
People who depend more on the CPUs
will get Core 2 Quad or Core2 Duo.

Let's not play replace the statement.

At least 23% of worldwide buyers bought AMD in Q3. It should have been 43%.

All hail the duopoly! QFX. Nipping on the heels of Kentsfield.
😳
 
And in case you didn't know the last stages of speed optimization for Windows come after the last RC.

Uh, No. When it's time to RC a product, especially one of that magnatude, you go into feature lock and fix only bugs. Optimizations after the last RC would at least paritally undermine the benefit of the RC releases.
 
mhz don't matter anymore

Clock per clock Core 2 spanks K8.

Nobody wins with 4x4 because it doesn't compete with anything, it just sucks. It won't sell.
Ok just underclock your CPU and save on power bills.

Clock per clock isnt fair because Intel doesnt have a C2Q clocked at 3GHz. :wink: :!:

I think I stated its a bad return on investment in the short term.
 
And in case you didn't know the last stages of speed optimization for Windows come after the last RC.

Uh, No. When it's time to RC a product, especially one of that magnatude, you go into feature lock and fix only bugs. Optimizations after the last RC would at least paritally undermine the benefit of the RC releases.

Well, as a person who shipped 5 versions of Windows, I guess I can say you don't know anything. Post-RC1 is used for last minute issues with some optimization and post-RC2 cleans up all debug and optimizes any possible kernel functions. Any reviewer can tell you that Vista Business RTM is MUCH faster and more responsive/polished than even RC2.

Check out Paul Thurott. Or maybe I should say Google Paul Thurott.
 
I only said salvage because if I had said improve there would have been a different discussion. I don't think it needs any help. People who want lower power will NOT GET DX10 SLI.

People who depend more on the CPUs won't get high end single GPUs. Voila, power is down.

It's a SERIOUSLY POWERFUL platform and tweaking will only make it better. As DDR CPUs disappear, DDR2 memory will drop in price, so by the time Vista hits higher clocks will be accessible to most people.
Getting the power down is important but it will require both a better 4X4 and 65nm FX's. Why would anyone buy and FX-72 or FX-74 when you could buy an FX-70 OEM and OC it to 3GHz? I think AMD made a misstake here marketing wise.

I would say Intel, by summer, will come with their own version of 4X4 and finish pushing the consumer into a server level upgrade path.
 
I will think about 4X4 if Barcelona is as good as it claims to be and when HT3 comes out..
This is simply worst time to consider 4X4.
 
I dont agree with BM because I dont think AMD has to salvage QFX for any chipset. Lets look at what AMD set out to do with 4X4. Beat C2D and even the lowest FX-70 does just that at a lower current price than the top C2D. Did 4X4 loss against the C2Q? Its a mobo so it really must be judged for what it was designed. It offers about the same GHz per GHz increase as C2Q with C2Q getting the best percentage on games and the 4X4 on multitasking.

What are the downfalls of both implementation of quad core? The 4X4 increases memory latency and costs more on the motherboard side. The C2Q6700 losses about 600MHz off its top OC over C2D's 4GHz and about 370MHz off its top released C2D6800.

What are the gains for both implementations of quad core? The FX-74 gains in refinement to give an added 200MHz over AMD's top FX-62. The 4X4 has a better upgrade path as its set for 2X-Agena FX's. The C2Q motherboard is cheaper than the 4X4 and one isnt forced to upgrade if Agena FX's performance is worse than expected.

Which company losses here? Both companies loss because this is a tech war. Intel losses due to having to modify a core well short in the cores life cycle and its top C2D6800 falling to 6th over all in about 4 months. AMD losses because the 4X4 and FX's will cut into its opty sales and could be a bad return on investment in the short term.

Which company wins? Neither because only we the consumer have won with cheaper faster CPU's.

You bring up a good point here --- if it were just QFX against the C2D dual core line up then the QFX is much more appealing alternative. This is a good example of timing as typical compares are best against best. Certianly, in high threaded situations, QFX will out do an X6800. This was probably AMD's original plans, and why they felt 'cheated' and that the quad core C2D was rushed.... or they could have thought, darn --- Intel will have a quad in Q4, we had better do somthing.... I kinda liked Anand's take on that.

In the long run, I see the 4x4 as a really good platform for AMD --- it is unfortunately for them that they had no alternative but to force a current chip/architecture into the form factor to get it to market --- they really need the K8L at this point, it make this particular product launch look like they are chasing Intel around (which, frankly, they are).

Nice post, thanks for the inputs.

Jack
That true and AMD does like to shake out new tech problems with old near perfect working wise parts. If Agena FX is a good performance CPU I would like to see Intel with a 4X4 like motherboard so this tech war can go on. Intel and AMD needs a 2 socket design that fixes the latency and I only see them getting 1 by pushing each other.

I think AMD is keeping the pressure on Intel until they get to K8L for their stock prices justification. Without the 4X4 Intel could of had a very profitable year and AMD having a flat year would have killed their stock.
 
mhz don't matter anymore

Clock per clock Core 2 spanks K8.

Nobody wins with 4x4 because it doesn't compete with anything, it just sucks. It won't sell.
Ok just underclock your CPU and save on power bills.

Clock per clock isnt fair because Intel doesnt have a C2Q clocked at 3GHz.

I think I stated its a bad return on investment in the short term.


That's where I disagree with you. It's not a bad short-term investment because ONLY Games show lower perf (some load up to 2GB). This is a platform designed to operate optimally in multi-core mode. Vista enables this for the mainstream desktop ( though it costs a BIT much - geez Bill it will take more than 50 years at this rate)

Anyway, I will let you know if my electric bill goes up significantly. Vista is due in Feb and R600 is due to at least be benched before then. Then I can objectively calculate my investment.

One site ( in the 4x4 reviews post) has little customizability and tops at more than you would care to know, but ibuypower.com has a slightly more flexible cofig that can cost as little as $2200 ( the approximate price of my 500W 4400+ system purchased almost exactly one year ago).

BTW, 800GTS SLI only requires 450W + 125-140W for the extra chip (AT 100% CPU AND GPU load). 600W PSUs are the mainstream now so it's not like it contains a boutique power supply.
 
Lie #1
I don't give out advice, dude.

Im not even going to waste time linking to your repetitive "C2D is like a car that goes 170, AM2 is like a car that goes 150....." , or your ever present "you can buy C2d, but in my opinion, AM2 would be a much better value...." shpeels. Everyone whose been on the board more than 2 weeks knows what Im talking about, and knows I am stating fact.

Guess what, anytime you recommend a product to some whos looking to buy something, it constitutes: .......whats that word.....you can say it.....thats right..... ADVICE.

Lie #2
This was never an AMD fanboy Forum. Intel followers were always holier-than-thou (insert colorful adjective) s.

Just like THG is a paid off Intel advertising site :roll: :

http://www.tomshardware.com/2006/02/21/a_look_at_amds_socket_am2_platform/page13.html#conclusion

http://www.tomshardware.com/2005/12/28/intels_65_nm_process_breathes_fire_into_double_core_extreme_edition/page25.html

http://www.tomshardware.com/2005/07/25/the_athlon_64_fx_overclocked_to_3_ghz/page10.html


You have been told time and time again, by the likes of Verndewd, myself and a dozen or more others that we all like and support AMD. Since we are members of this forum, it cannot be an Intel "holier than thou" fanboy forum.

The lie here is not about AMD, it is about you. And most disturbingly, the lie is not to us, but to yourself. The lie is about support and forum prefernces. No one supports or prefers is your mis-interprtation of information, presentation of fiction as fact, or statement of opinion as fact. The actual lie is that you equate a lack of support for you and your propaganda directly to a lack of support for AMD. You are wrong. The reality you fail to grasp is that you and AMD are not the same entity and that much more often than not, you are significantly wrong (through lack of understanding) about their products performance attributes, both current and future.

Personality Defect #1
That's why I don't give any of you the benefit of the doubt. This is just for fun. It is not for edification (I hope) or advanced CPU theory.

That you are either claiming to, or are in reality having fun being the butt of the joke and the recipient of endless abuse/debunking is indicative of an emotional problem of significant proportions. Seek help immediatly

At a loss for a description #1(self description?)
Just for fun. I stll plan on buying the platform, regardless of what Intel has out. If I'm dumb for it, fine, I will never ask you to use the PC.

If????? Did you say IF? At this point in time, I will wager that the general concensus is that there's no longer any question.

Clueless #1
I mean why are you trying to tell me how to spend my money? I am paying so I buy what I want, not what the denziens of a public forum want.
Yes, C2Q is faster but both are faster than my 4400+ and I prefer AMD.

Once again, because you keep offering the information, in a public form, unsolicited. I have yet to see anyone post..."Hey Baron, what are you buying?"
You offer your opinions, you offer your advice, and you offer your purchasing plans (in what I can only interpret as some deluded attempt to validate the product) with no prompting what so ever from anyone.
As such, you have no right to complain, object or whine (as you do persistantly) when anyone responds negatively to your input. Once again, if you dont want to hear about how a fool and his money are soon parted, stop telling us how you plan to part with your money.

Flawed Logic #1
Again I find it amazing that no one catches on to the fact that people who buy this will be upgrading. This means that the average purchaser who has 939 or AM2 will get 60-200% perf increases, depending on where they started which FX and which app.

This is your opinion Baron. It is in no way, shape or form fact. I have yet to see a single article or review which recommends let alone mentions spending $2500 and up just for the capability to upgrade. While I have seen mention that 4x4 will be upgradable, this is not the same as saying "you should buy it now so you can upgrade it later." Spending that kind of money for the sole capability of upgrading to a product (AMD Quadcore) which does not yet exist is another example of galactically moronic logic.

The fact that you took mention of the capability of upgrading to quadcores and "developed" the upgrade-purchase argument and have pushed it so much that you now actually believe it to be a valid argument makes this a multi faceted problem: Flawed logic #1 and Personality Defect #2...that being you believe your own bullshite.

Personally, I find it amazing that anyone would even consider buying a brand new, outrageously expensive, unproven platform/system, based on the premise "I can stick quad cores in it later" Maybe a business, for servers, but then only with a proven platform. No one, and I mean no one, with an ounce of intelligence, will willingly risk such a critical and expensive descision on un-validated hardware. The one caveat to that is those who wish to test/evaluate the system, and they are not risking a companies income on that.

At a loss for a description #2 (contradiction)
I can't wait for the first Vista multithreaded game to come out.

Yet from an earlier post we have:
You're an idiot. I am NOT , I REPEAT NOT buying this for games. It will do well in the multithreaded games but that's wat the DX10 GPU is for. The CPUs are for Visual Studio, SQL Server and Virtual Server.

These chips give me four cores of Virtualization. That's what I want.

I guess you judt shouldn't follow my purchase needs thugh, since the Intel club would drum you out.

Any of the 12lb brains care to jump on this one? "I Cant wait for the first Vista games but I am not, I repeat NOT buying this for games"

The best I can figure with this one is to anticipate the "BaronBackPeddle" ---"....um..uh...err...ahh..what I really meant was Im buying it for "dev"ing, but I also plan to play games. Uh yeah, thats what I meant....not that I wasnt going to play games even though I emphasized that I wasnt buying it for games."

At a loss for a description #3 (pointless?)
QFX will smoke. C2Q will smoke. Also, for those who say I'm giving people advice, how is that any different than you trying to talk me out of my purchase?

On this we can agree. So much so in fact, I believe AMD should ship the below item as a complementary gift with every purchase of QFX.

engine1.jpg


Not that I believe OFX will actually catch fire...just when its power draw causes your house wiring to heat up and ignite the wall studs/framing, the resultant fire can be extinguished before it reaches and harms the capitol investment that QFX is.

Excuse making #1
I guess we'll have to wait R600 and the RD600 chipset to see if my questions yields a positive. I think it will since NUMA is designed to put the memory nearest the chip.

Once again looking to a future product to save the performance of a brand new product in relation to a current product. How very Intel/Heatburstish of you. It is both outrageuosly funny and hypocritical that you vigerously cast aspersions at Intel, yet are not above using the very tactics they used to promote Netbust to try and "save face" for AMD. All the more so since all you actually accomplish is to further besmirch AMDs reputation.

Excuse making #2 (with lack of proof # 2853)
It's obvious that the issues reported by Legit Review (IIRC) showed that it wasn't working so app data would get split amongst sockets.

Vista is said to improve this a lot while of course being based on multi-core. If they keep the OS in one socket, most apps will always fit in to the other banks. 2GB is a large process size.
Some complex apps WILL use sets that size but they are already multithreaded and perf should be nearly identical to the Opteron 2xxx with the same clock and amount of RAM.

Sigh.....Once again the phase "its said" pops up with no evidence to support the statement. How very BaronMatrix of you Baron.

While I have no doubt Vista will aid the QFX (only to steal that performance back to support its own sub-routine bloatware) you are again looking to another product to solve the problems of a new product.

In the real world, we call this...."making excuses"

All hail QFX.
[/quote]

All hail the NYC Dept of Heath, Mental Health Division, and dispacth them to BMs locale . Most Ricky Tick.
 
I dont agree with BM because I dont think AMD has to salvage QFX for any chipset. Lets look at what AMD set out to do with 4X4. Beat C2D and even the lowest FX-70 does just that at a lower current price than the top C2D. Did 4X4 loss against the C2Q? Its a mobo so it really must be judged for what it was designed. It offers about the same GHz per GHz increase as C2Q with C2Q getting the best percentage on games and the 4X4 on multitasking.

What are the downfalls of both implementation of quad core? The 4X4 increases memory latency and costs more on the motherboard side. The C2Q6700 losses about 600MHz off its top OC over C2D's 4GHz and about 370MHz off its top released C2D6800.

What are the gains for both implementations of quad core? The FX-74 gains in refinement to give an added 200MHz over AMD's top FX-62. The 4X4 has a better upgrade path as its set for 2X-Agena FX's. The C2Q motherboard is cheaper than the 4X4 and one isnt forced to upgrade if Agena FX's performance is worse than expected.

Which company losses here? Both companies loss because this is a tech war. Intel losses due to having to modify a core well short in the cores life cycle and its top C2D6800 falling to 6th over all in about 4 months. AMD losses because the 4X4 and FX's will cut into its opty sales and could be a bad return on investment in the short term.

Which company wins? Neither because only we the consumer have won with cheaper faster CPU's.

Interesting perspective and a very logical argument given the timing, though I would disagree (respectfully of course 😀 ) on the upgrade path option for AMD. Due to the wafer usage and binning, I dont think AMD will be able to price Agena agressively enough to make selling them in box sets of 2 an attractive option for all but a select few consumers with excessive money to burn.
 
I only said salvage because if I had said improve there would have been a different discussion. I don't think it needs any help. People who want lower power will NOT GET DX10 SLI.

People who depend more on the CPUs won't get high end single GPUs. Voila, power is down.

It's a SERIOUSLY POWERFUL platform and tweaking will only make it better. As DDR CPUs disappear, DDR2 memory will drop in price, so by the time Vista hits higher clocks will be accessible to most people.
Getting the power down is important but it will require both a better 4X4 and 65nm FX's. Why would anyone buy and FX-72 or FX-74 when you could buy an FX-70 OEM and OC it to 3GHz? I think AMD made a misstake here marketing wise.

I would say Intel, by summer, will come with their own version of 4X4 and finish pushing the consumer into a server level upgrade path.

Cool n Quiet got the FX74 down to the same idle power as C2Q. I don't think singlethreaded games will really get a big increase with other than clock-speed. Though in a theoretically-correct NUMA environment, swapping has precedence over sharing; such that a 2GB process can take longer to load while swapping rather than placing significant amounts of data on the other socket's RAM banks.

Raising the RAM to 8GB in 4 sockets would actually eclipse Opteron becasue of the speedier non-ECC RAM. It has been "Inq-reported" that the Agena version will support 8GB to keep the differentation. This is an indicator (along with the 1.9GHz clock speed) that Barcelona will be a monster.

A 70% fatser 2.3GHz QFX would be almost unholy, especially if the "Inq-reported" HT2 claims for it are unfounded.
 
Cool n Quiet got the FX74 down to the same idle power as C2Q.

Could you link that data please.

Power consumption of a Quad FX system is simply unreal for a desktop, as it should be because this is effectively a workstation platform with un-buffered memory. At idle our Quad FX test bed consumed nearly 400W, partially because we couldn't get Cool 'n Quiet running on the system, but also because the CPUs and motherboard simply draw an incredible amount of power. Update: We got Cool 'n Quiet working on the motherboard which reduced idle power significantly, down to within a few watts of the Kentsfield system. Load power was unchanged.

CPU Idle Power Load Power Performance per Watt (fps/watt)
AMD Athlon 64 FX-74 (3.0GHz x 4) 217W 456W 17.7
Intel Core 2 Extreme QX6700 (2.66GHz x 4) 213W 263W 32.9


Looking at power consumption under full load, Cool 'n Quiet would have no chance to even make an impact as all cores are being utilized at full speed. Under load the Quad FX system pulled 456W on average, a full 73% more than our Kentsfield testbed.

whatever
 
mhz don't matter anymore

Clock per clock Core 2 spanks K8.

Nobody wins with 4x4 because it doesn't compete with anything, it just sucks. It won't sell.
Ok just underclock your CPU and save on power bills.

Clock per clock isnt fair because Intel doesnt have a C2Q clocked at 3GHz.

I think I stated its a bad return on investment in the short term.


That's where I disagree with you. It's not a bad short-term investment because ONLY Games show lower perf (some load up to 2GB). This is a platform designed to operate optimally in multi-core mode. Vista enables this for the mainstream desktop ( though it costs a BIT much - geez Bill it will take more than 50 years at this rate)

Anyway, I will let you know if my electric bill goes up significantly. Vista is due in Feb and R600 is due to at least be benched before then. Then I can objectively calculate my investment.

One site ( in the 4x4 reviews post) has little customizability and tops at more than you would care to know, but ibuypower.com has a slightly more flexible cofig that can cost as little as $2200 ( the approximate price of my 500W 4400+ system purchased almost exactly one year ago).

BTW, 800GTS SLI only requires 450W + 125-140W for the extra chip (AT 100% CPU AND GPU load). 600W PSUs are the mainstream now so it's not like it contains a boutique power supply.
4X4 is bad short term because it will come cheaper and need less power with revisions. I see this occuring by the time AMD rolls out 65nm in volume.
 
I think AMD is keeping the pressure on Intel until they get to K8L for their stock prices justification. Without the 4X4 Intel could of had a very profitable year and AMD having a flat year would have killed their stock.

This is where I will respectfully disagree -- it isn't Intel currently underpressure from a performance prespective.... and the marketshare trend is turning around.

http://www.economist.com/business/displaystory.cfm?story_id=8355132


While the MSS trend has been in AMD's favor, and the pressure was on Intel in the last quarter of 2005/1H 2006, it would appear they have turned the corner.... C2D has, so far, caused AMD to drop prices on the DT > 60%, offer rebates on most all remaining stock, and basically rush out a 4x4 solution.

Now, this is just what it appears --- it may trend off lower or trend off higher. But, after studying the dynamic of the MSS for DT, Sever, and mobile it is beginning to look like Intel's MSS trend is turning around. AMD did not take share from Intel in Q3, and if you scrub for the Via affect, Intel gained in both sever and DT (AMD had a great mobile showing however). It will take a few quarters to see how much pressure comes from one side or the other.... but AMD is currently, from a technical perspective, on the defensive.

I would go futher to say that they are (as they have been) on the manufacturing capacity/QC defensive as well. Its fun to watch and participate in the "market share" game, but ultimately share size today means little for tommorrow as the market is not stagnent in terms of size. It is, has been and will continue to grow in all segments (albeit at difrerent rates), and AMD has to face the challenge that as the market continues to grow, they may lose a percentage of shares as their production capacity peaks. When they can no longer expand at the same rate as the market (in reality I think we are already seeing AMDs production capacity at its peak) they simply will not be able to retain, let alone expand their current share. This of course means nothing in terms actual numbers produced for them, but as the market continues to expand and they begin to loose percentage, one of the chief results will be the lose of the all important PR "prestige" they gained earlier this year. With the NY fab not set to open for several years, they desperately need to get 65nm up to speed just to reduce die size and increase wafer capacity, let alone worry about any power and perfromace gains 65nm will bring them.
 
I dont agree with BM because I dont think AMD has to salvage QFX for any chipset. Lets look at what AMD set out to do with 4X4. Beat C2D and even the lowest FX-70 does just that at a lower current price than the top C2D. Did 4X4 loss against the C2Q? Its a mobo so it really must be judged for what it was designed. It offers about the same GHz per GHz increase as C2Q with C2Q getting the best percentage on games and the 4X4 on multitasking.

What are the downfalls of both implementation of quad core? The 4X4 increases memory latency and costs more on the motherboard side. The C2Q6700 losses about 600MHz off its top OC over C2D's 4GHz and about 370MHz off its top released C2D6800.

What are the gains for both implementations of quad core? The FX-74 gains in refinement to give an added 200MHz over AMD's top FX-62. The 4X4 has a better upgrade path as its set for 2X-Agena FX's. The C2Q motherboard is cheaper than the 4X4 and one isnt forced to upgrade if Agena FX's performance is worse than expected.

Which company losses here? Both companies loss because this is a tech war. Intel losses due to having to modify a core well short in the cores life cycle and its top C2D6800 falling to 6th over all in about 4 months. AMD losses because the 4X4 and FX's will cut into its opty sales and could be a bad return on investment in the short term.

Which company wins? Neither because only we the consumer have won with cheaper faster CPU's.

Interesting perspective and a very logical argument given the timing, though I would disagree (respectfully of course 😀 ) on the upgrade path option for AMD. Due to the wafer usage and binning, I dont think AMD will be able to price Agena agressively enough to make selling them in box sets of 2 an attractive option for all but a select few consumers with excessive money to burn.
Why do think you must buy a box set? Agena FX may work with only 1 CPU on some 4X4's. Technology that grows with your needs comes to mind.
 
I dont agree with BM because I dont think AMD has to salvage QFX for any chipset. Lets look at what AMD set out to do with 4X4. Beat C2D and even the lowest FX-70 does just that at a lower current price than the top C2D. Did 4X4 loss against the C2Q? Its a mobo so it really must be judged for what it was designed. It offers about the same GHz per GHz increase as C2Q with C2Q getting the best percentage on games and the 4X4 on multitasking.

What are the downfalls of both implementation of quad core? The 4X4 increases memory latency and costs more on the motherboard side. The C2Q6700 losses about 600MHz off its top OC over C2D's 4GHz and about 370MHz off its top released C2D6800.

What are the gains for both implementations of quad core? The FX-74 gains in refinement to give an added 200MHz over AMD's top FX-62. The 4X4 has a better upgrade path as its set for 2X-Agena FX's. The C2Q motherboard is cheaper than the 4X4 and one isnt forced to upgrade if Agena FX's performance is worse than expected.

Which company losses here? Both companies loss because this is a tech war. Intel losses due to having to modify a core well short in the cores life cycle and its top C2D6800 falling to 6th over all in about 4 months. AMD losses because the 4X4 and FX's will cut into its opty sales and could be a bad return on investment in the short term.

Which company wins? Neither because only we the consumer have won with cheaper faster CPU's.

Interesting perspective and a very logical argument given the timing, though I would disagree (respectfully of course 😀 ) on the upgrade path option for AMD. Due to the wafer usage and binning, I dont think AMD will be able to price Agena agressively enough to make selling them in box sets of 2 an attractive option for all but a select few consumers with excessive money to burn.
Why do think you must buy a box set? Agena FX may work with only 1 CPU on some 4X4's. Technology that grows with your needs comes to mind.

Another interesting point, but if you only want to upgrade to one CPU, why bother buying a 4x4 in the first place?