Phenom B3 stepping black edition full Reveiw @ FiringSquad

[Idontcare]

Originally posted by: Viditor

Originally posted by: aigomorla
i just cant see how AMD expects to compete against the Q6600 even.

And even intels own Q9450 series is having a hard time catching up to a Q6600 G0 in overclocking numbers.

Which leads me to think, AMD has issues or is best to be aiming in a OEM market. But i still think they would get killed by a kentsfield in that market even.

You are speaking of overclockers, which is less than 1% of the market...I really don't think AMD is too worried about that.
Where the B3 Phenoms should do exceedingly well is in the sweet spot (mainstream sales).

I actually saw a report where Intel stated they estimate the overclocking/enthusiast community to represent ~5% of the market. It did not say whether that was $ volume or chip colume.

This was last fall sometime. I don't have the link but the number was burned into my brain because I couldn't beleive it was actually so large. I still have a hard time beleiving it but the number is out there per Intel. Could have been a fluke I suppose, maybe they already fired that guy for the typo

If I find the link I'll post it.

 

[Viditor]

Originally posted by: Idontcare

Originally posted by: Viditor

Originally posted by: aigomorla
i just cant see how AMD expects to compete against the Q6600 even.

And even intels own Q9450 series is having a hard time catching up to a Q6600 G0 in overclocking numbers.

Which leads me to think, AMD has issues or is best to be aiming in a OEM market. But i still think they would get killed by a kentsfield in that market even.

You are speaking of overclockers, which is less than 1% of the market...I really don't think AMD is too worried about that.
Where the B3 Phenoms should do exceedingly well is in the sweet spot (mainstream sales).

I actually saw a report where Intel stated they estimate the overclocking/enthusiast community to represent ~5% of the market. It did not say whether that was $ volume or chip colume.

This was last fall sometime. I don't have the link but the number was burned into my brain because I couldn't beleive it was actually so large. I still have a hard time beleiving it but the number is out there per Intel. Could have been a fluke I suppose, maybe they already fired that guy for the typo

If I find the link I'll post it.

I'd be dubious about that as well...maybe 5% of the US consumer market. That should be far less than 1% of the total market. Do look though...I've been very wrong before.

 

[CTho9305]

Originally posted by: Idontcare

Originally posted by: ExarKun333

Originally posted by: CTho9305

Originally posted by: ExarKun333
God the Phenom still looks like crap. Everyone is reviewing these and stating they are so much better than the B2s, when all that has been fixed is the TLB issue.

Another review by Hothardware illustrates the power consumption of the Phenom 2.5 (load) @ 346 and the Q6600 (stock-load) @ only 255 (I am not really even going to mention the QX9650 being 206).

This reminds of me when the P4E's were using 2x as much power and every person and their brother complained about the power usage and heat.

I built by first Intel rig EVER last year and always loved AMD (still do) but the Phenom is still crap. It is no wonder they won't up the clockspeed on these guys; I would be curious to see the power usage at 2.8 and 3.0.

edit:SP


Hothardware - System Power Consumption

Why would idle power consumption differ for B2 with the TLB patch enabled vs disabled? If the system is idle, why would it matter what the TLB does? Hot Hardware shows a 6 watt difference... yet the load power didn't really change. That makes absolutely no sense to me. The load power, for a real application, should be lower with the TLB patch enabled (because the CPU spends more time waiting for data from memory), right?

I wonder if power is still being used in the L3 cache with the TLB patch enabled? Please dont flame me for this remark, just speculation on my part. Maybe the cpu can manage the power better with the cache enabled and either use or reduce power as needed. Anyone have more information?

CTho9305 is right, you are thinking about it backwards. The TLB patch basically forces data to the L3 instead of staying resident in the L2...kind of like turning off the L2 but not nearly so severe.

So L3 is powered on and if anything getting updated more than usual...but because the individual cores themselves ought to be idling more often with the TLB patch than without the patch (that is what causes the lower performance) the overall average power consumption ought to be lower for the B2 patched system.

CTho9305,
As to the 6W delta...I'd call that noise. The PSU efficiency itself could be meandering a mere 6W and still be well within specs. 20W (10%) would raise my eyebrow, but not a mere 6W discrepancy.

That seems like it generally invalidates a lot of power measurements review sites do - especially when evaluating lower-power processors. 6 W of noise is huge when you're talking about a 65W CPU, and it's still significant when you're talking about 89/95W CPUs. At the very least, it means sites shouldn't draw conclusions about power consumption based on their very-small-sample-size measurements.

 

[bryanW1995]

is that 6w on the cpu only or 6w on the total system?

re the 5% total market for the enthusiast community, that is probably by dollar volume and does make sense as we are much more apt to throw money at hardware just for the fun of it (see markfw900 and aigomorla for examples of "throwing money at hardware")

 

[Idontcare]

Originally posted by: CTho9305

Originally posted by: Idontcare
CTho9305,
As to the 6W delta...I'd call that noise. The PSU efficiency itself could be meandering a mere 6W and still be well within specs. 20W (10%) would raise my eyebrow, but not a mere 6W discrepancy.

That seems like it generally invalidates a lot of power measurements review sites do - especially when evaluating lower-power processors. 6 W of noise is huge when you're talking about a 65W CPU, and it's still significant when you're talking about 89/95W CPUs. At the very least, it means sites shouldn't draw conclusions about power consumption based on their very-small-sample-size measurements.

Actually what invalidates 99% of published power measurement reviews is the fvcking sample size is ONE.

I can't over-stress how much of a pet peeve this is of mine. It drives be beyond crazy. How freaken lazy do you have to be as a review site to not acquire 5-10 samples of the same SKU, drop them into a testbed and calculate some average and standard deviations on power consumption?

But yes, along with single samples undermining the utility of power measurements they typically only report a single "run" value. I so badly miss the days of THG when the Pabster was in charge and he'd run all benches in triplicate to report avg +/- 1-sigma.

I tested multiple Q6600 systems with the same kill-a-watt device and power consumption varies by at least 10W across my 5 samples. Within the same system it can vary 5-10W during a full load scenario.

But seriously the laziness of the reviewers is ridiculous. This holds true for GPU power consumption reviews. And overclocking reviews for both CPU and GPU. Get 5 samples and do it right.

 

[batmang]

If your not overclocking, I still think the Phenom's are a damn good deal. They perform well at a low price. Its a nice tradeoff. Your not paying more and your getting Q6600-like performance. Whats not to like?

If your a overclocking nut, you would be a complete dummy to get a Phenom. Thats all there is to it. Since I'm not a overclocking freak and I'm a lover of ATi, I went with the Spider platform. 790FX, HD3870, Phenom 9600, the whole deal. I absolutely love it. It does everything I expected it to do. What makes me laugh about this whole " AMD is crap " mentality is the AMD haters are not seeing the flaw to this. Compeition drops pricing.
I hope ATi releases a 9800GX2 killer, I hope the 45nm Phenom's clock much higher and have overclocking head room. I want Intel to stay uncomfortable. It makes the products affordable. A perfect example is the 9600GT and 3870's, look how cheap they are!

 

[lopri]

I doubt 45nm Phenom will clock much higher. AMD's design and choice of material have never been for high clocks. Intel, on the other hand, have always had the know-how of high frequencies. AMD has always competed on better performance per clock, not high clocks. While Intel has hit a home-run with Conroe, I still think Phenom is a very 'elegant' design which unfortunately is ahead of its time (meaning not many apps can take advantage of its strength). When I see a mature platform for Phenom I will definitely try it out.

 

[golem]

Originally posted by: lopri
I doubt 45nm Phenom will clock much higher. AMD's design and choice of material have never been for high clocks. Intel, on the other hand, have always had the know-how of high frequencies. AMD has always competed on better performance per clock, not high clocks. While Intel has hit a home-run with Conroe, I still think Phenom is a very 'elegant' design which unfortunately is ahead of its time (meaning not many apps can take advantage of its strength). When I see a mature platform for Phenom I will definitely try it out.

I've read Phenoms do well in the server area, but how could desktop apps be changed to run better on the Phenom?

 

[Idontcare]

Originally posted by: lopri
I doubt 45nm Phenom will clock much higher.

Are you talking about round1 45nm (SiON/PolySi xtors + Low-k interconnect) or round2 45nm (Hi-k/MG xtors + ultra-Low-k interconnect)?

I agree if you restricting this to a round1 45nm discussion.

 

[Ratman6161]

Another way to look at this though is that AMD does in fact have the lowest priced quad core cpu available in the 9550. All the reviews we have been seeing have only tested the 9850's but I'm wondering if a 9550 would overclock just as high as a 9850? At $40 less than a 9850, this might change the price/performance picture a little bit. All that said, and as much as I have been an AMD fan for at least 10 years, about the only way I can see one of these being viable for an enthusiast right now is if you already have a decent AM2 board that would accept a phenom as a drop-in replacement for your A64 X2. In my case I was still on socket 939 and so the Q6600 was the only way to go for me..

 

[classy]

The Phenoms appear to be not that bad. It looks like in most cases you probably won't notice a difference compared to Intel cpus. They are a tough sell because of less performance but not too bad at all.

 

[ArchAngel777]

I have always been a bit partial for AMD, mostly because their prices were reasonable. However, I cannot help but think that the Phenom is a complete failure. I just don't see it any other way. When I checked out the review that Firing Squad did, I just shook my head because the performance, at least in gaming, was so sub-par.

Will you notice the difference? Maybe, it depends. You will certainly notice it in the future as GPU's become more powerful. I hope that AMD can produce a better revision here in the future.

 

[Phynaz]

Originally posted by: Idontcare

Originally posted by: lopri
I doubt 45nm Phenom will clock much higher.

Are you talking about round1 45nm (SiON/PolySi xtors + Low-k interconnect) or round2 45nm (Hi-k/MG xtors + ultra-Low-k interconnect)?

I agree if you restricting this to a round1 45nm discussion.

It may also apply to round 2. Note that IBM has not shipped a Power CPU made on high-K. I wonder why?

 

[lopri]

Originally posted by: Idontcare
Are you talking about round1 45nm (SiON/PolySi xtors + Low-k interconnect) or round2 45nm (Hi-k/MG xtors + ultra-Low-k interconnect)?

I agree if you restricting this to a round1 45nm discussion.

I honestly don't have that much technical knowledge but I was thinking of 65nm SOI and 45nm SOI from AMD in general.

 

[Idontcare]

Originally posted by: lopri

Originally posted by: Idontcare
Are you talking about round1 45nm (SiON/PolySi xtors + Low-k interconnect) or round2 45nm (Hi-k/MG xtors + ultra-Low-k interconnect)?

I agree if you restricting this to a round1 45nm discussion.

I honestly don't have that much technical knowledge but I was thinking of 65nm SOI and 45nm SOI from AMD in general.

AMD has publicly stated there are two stages (tier's if you will) to their 45nm node.

The first stage will essentially be a standard shrink with no new materials introduction relative to the 65nm tech (still SOI, standard silicon oxynitride gate oxide, standard doped polysilicon gates, standard blackdiamond low-k interconnect dielectric).

Then "sometime later" AMD will release a second stage of 45nm tech which will include Hik/MG transistors and ULK blackdiamond-II for interconnects (reduces capacitance, which reduces parasitic power consumption and reduces wire delay).

I could see the initial 45nm stuff offering nothing more than a cheaper die (smaller, thus more per wafer) for AMD. Some power reduction and some voltage reduction, which could both be consumed by upping the clocks and gaining a speedbin or two.

But when the hik/mg with ULK come out it had better be higher clocked and lower power or it will really be a technology failure.

 

[Viditor]

Originally posted by: Idontcare

Originally posted by: lopri

Originally posted by: Idontcare
Are you talking about round1 45nm (SiON/PolySi xtors + Low-k interconnect) or round2 45nm (Hi-k/MG xtors + ultra-Low-k interconnect)?

I agree if you restricting this to a round1 45nm discussion.

I honestly don't have that much technical knowledge but I was thinking of 65nm SOI and 45nm SOI from AMD in general.

AMD has publicly stated there are two stages (tier's if you will) to their 45nm node.

The first stage will essentially be a standard shrink with no new materials introduction relative to the 65nm tech (still SOI, standard silicon oxynitride gate oxide, standard doped polysilicon gates, standard blackdiamond low-k interconnect dielectric).

Then "sometime later" AMD will release a second stage of 45nm tech which will include Hik/MG transistors and ULK blackdiamond-II for interconnects (reduces capacitance, which reduces parasitic power consumption and reduces wire delay).

I could see the initial 45nm stuff offering nothing more than a cheaper die (smaller, thus more per wafer) for AMD. Some power reduction and some voltage reduction, which could both be consumed by upping the clocks and gaining a speedbin or two.

But when the hik/mg with ULK come out it had better be higher clocked and lower power or it will really be a technology failure.

A couple of points here IDC...
1. The first version will also have a new type of Si Strain
2. The 2nd version is to be released in Q1 09

 

[Extelleron]

Originally posted by: Viditor

Originally posted by: Idontcare

Originally posted by: lopri

Originally posted by: Idontcare
Are you talking about round1 45nm (SiON/PolySi xtors + Low-k interconnect) or round2 45nm (Hi-k/MG xtors + ultra-Low-k interconnect)?

I agree if you restricting this to a round1 45nm discussion.

I honestly don't have that much technical knowledge but I was thinking of 65nm SOI and 45nm SOI from AMD in general.

AMD has publicly stated there are two stages (tier's if you will) to their 45nm node.

The first stage will essentially be a standard shrink with no new materials introduction relative to the 65nm tech (still SOI, standard silicon oxynitride gate oxide, standard doped polysilicon gates, standard blackdiamond low-k interconnect dielectric).

Then "sometime later" AMD will release a second stage of 45nm tech which will include Hik/MG transistors and ULK blackdiamond-II for interconnects (reduces capacitance, which reduces parasitic power consumption and reduces wire delay).

I could see the initial 45nm stuff offering nothing more than a cheaper die (smaller, thus more per wafer) for AMD. Some power reduction and some voltage reduction, which could both be consumed by upping the clocks and gaining a speedbin or two.

But when the hik/mg with ULK come out it had better be higher clocked and lower power or it will really be a technology failure.

A couple of points here IDC...
1. The first version will also have a new type of Si Strain
2. The 2nd version is to be released in Q1 09

I doubt that AMD will launch 45nm CPUs w/o HK+MG in Q4 '08, and then 3 months later, release CPUs with HK+MG.

I would expect AMD will move to their "second tier" 45nm process in the middle of 2009, if at all. Something tells me that AMD will delay HK+MG until 32nm. If they do implement HK+MG on 45nm, it will likely be when AM3 CPUs are released.

 

[Viditor]

Originally posted by: Extelleron

I doubt that AMD will launch 45nm CPUs w/o HK+MG in Q4 '08, and then 3 months later, release CPUs with HK+MG.

I would expect AMD will move to their "second tier" 45nm process in the middle of 2009, if at all. Something tells me that AMD will delay HK+MG until 32nm. If they do implement HK+MG on 45nm, it will likely be when AM3 CPUs are released.

AMD is ramping 45nm as we speak, and they are due to begin shipping this summer.
AM3 is due to ship in Q1 09...

Edit: The AM3 based Montreal chips should have availability in Q2 09...

 

[Idontcare]

Originally posted by: Extelleron
I doubt that AMD will launch 45nm CPUs w/o HK+MG in Q4 '08, and then 3 months later, release CPUs with HK+MG.

I would expect AMD will move to their "second tier" 45nm process in the middle of 2009, if at all. Something tells me that AMD will delay HK+MG until 32nm. If they do implement HK+MG on 45nm, it will likely be when AM3 CPUs are released.

I'd agree if these were ordinary business conditions...but AMD is a desperate animal which has been cornered and wounded...nothing so dangerous as management willing to take on riskier and riskier decisions as every day passes. (look how Intel's Core architecture was born...)

The CIT culture at AMD makes it all the more viable to pull Hik/MG into 45nm as it comes mature enough to warrant doing so. It may rob some of the thunder from 32nm, as 90nm CIT's appear to have done to 65nm, but AMD needs profits sooner instead of later.

Remember the Thoroughbred-B where they managed to add a metal level and respin the 130nm Thoroughbred-A core in literally a matter of months?

<a target=_blank class=ftalternatingbarlinklarge href="http://www.anandtech.com/showdoc.aspx?i=1635&p=3">Thoroughbred-A reviewed at Anandtech June 10, 2002
</a>

<a target=_blank class=ftalternatingbarlinklarge href="http://www.anandtech.com/showdoc.aspx?i=1685&p=3">Thoroughbred-B reviewed at Anandtech August 21, 2002
</a>

That was crazy silly to do that on such a timeline, but they did because they needed to really come back hard and fast on Northwood. Never underestimate the desperation of a wounded and cornered animal.

 

[SickBeast]

Originally posted by: Idontcare
Never underestimate the desperation of a wounded and cornered animal.

AMD has been wounded for years now.

At this point they've been ravaged, mangled, and left for dead.

They do have a pulse, however, and are most certainly desperate at this point.

The thing is, the Phenom is actually fairly competetive at this point (and is actually winning in the server market). If intel hadn't released 45nm already, I'd say that AMD is actually doing fine.

 

[Idontcare]

Originally posted by: SickBeast

Originally posted by: Idontcare
Never underestimate the desperation of a wounded and cornered animal.

AMD has been wounded for years now.

At this point they've been ravaged, mangled, and left for dead.

They do have a pulse, however, and are most certainly desperate at this point.

The thing is, the Phenom is actually fairly competetive at this point (and is actually winning in the server market). If intel hadn't released 45nm already, I'd say that AMD is actually doing fine.

True, getting gang-ravaged by an 800lb gorilla (Intel) and a 400lb gorilla jr (Nvidia) will put you out of commission for a little while

Thankfully for Hector and his crack executive team's strategy of operation employee human shield has kept the corporate elites from getting anything mangled except their ego's in the fruckus since 2006.

 

[bryanW1995]

HEY, WAIT A MINUTE! Have you ever heard of the EMANCIPATION PROCLAMATION? - chef

"I don't listen to hip hop" - general

 

[Idontcare]

Originally posted by: bryanW1995
HEY, WAIT A MINUTE! Have you ever heard of the EMANCIPATION PROCLAMATION? - chef

"I don't listen to hip hop" - general

You read my mind. South Park FTW!

 

[nyker96]

Doesn't look like anyone got it pass 3ghz yet. would be nice if they can, I believe maybe the core is the problem that's limiting max OC so maybe they'll do much better at 45nm but that core needs to be updated to OC well.

 

[superstition]

AMD has always competed on better performance per clock

Hasn't the Core architecture eliminated that advantage, though, along with work per watt?