Does AMD have anything special on the horizon?

[harpoon84]

Viditor, geez, you're quite eager for K8L to reclaim the performance crown, aren't you?

If I get a dollar for each time you've proclaimed K8L to be the 'new performance leader' I'd make a tidy little forture.

Why do you keep stressing out whenever there is someone who doesn't quite share the same optimism as you do about AMD's prospects just by looking at paper specs.

There are just two many variables at the moment for anyone to be making a solid 'K8L will beat Core2' claim. We don't even have any leaked benches or ES for folks to test.

Even if K8L is faster than Core2 clock for clock, there is still the respective clockspeeds of the two architectures that need to be accounted for.

Right now, nobody knows for sure if K8L is faster than Core2 clock for clock, and if it is, whether it'll be clocked fast enough compared to 45nm Core2s to give it the performance crown.

You seem to have jumped to the conclusion that K8L will annihilate Core2 though, so either you've got an ES sample you're not telling us about , or you have the utmost faith in AMD to deliver as promised.

Would I be correct in assuming you are an AMD shareholder?

 

[Viditor]

Originally posted by: DrMrLordX

Intel will probably have a clock speed advantage if K8L tops out at 2.9 ghz on release. Probably. However, if Intel rolls out a stock 3.0-3.2 ghz Kenstfield to counter K8L, expect it to be a hot chip (literally).

Whether or not said clock speed advantage will be enough to tip scales in Intel's favor remains to be seen. I imagine that it will.

The thing is that Kentsfield won't be the competition, that would be Cloverton (remember that the first K8L Quad Cores will be Opterons...).
Frankly, I have no idea what the clockspeeds of either will be...actually nobody does (despite the rampant guessing). We have yet to see a 65nm AMD overclocking test yet, so when some people start predicting clockspeeds, they can't really be doing it based on facts...

 

[Viditor]

Originally posted by: harpoon84
Viditor, geez, you're quite eager for K8L to reclaim the performance crown, aren't you?

If I get a dollar for each time you've proclaimed K8L to be the 'new performance leader' I'd make a tidy little forture.

Haven't done it even once, mate... If you read again, I always say that it's IF AND ONLY IF things go the way that's been published. I was equally cautious when C2D was first announced...

Why do you keep stressing out whenever there is someone who doesn't quite share the same optimism as you do about AMD's prospects just by looking at paper specs.

There are just two many variables at the moment for anyone to be making a solid 'K8L will beat Core2' claim. We don't even have any leaked benches or ES for folks to test.

No stress at all... I absolutely agree that we only have paper specs, which means we can't make real conclusions EITHER WAY...
That said, I see many posters who make wild assumptions about what speeds are going to be, performance determinations, etc...

Even if K8L is faster than Core2 clock for clock, there is still the respective clockspeeds of the two architectures that need to be accounted for.

Right now, nobody knows for sure if K8L is faster than Core2 clock for clock, and if it is, whether it'll be clocked fast enough compared to 45nm Core2s to give it the performance crown.

You seem to have jumped to the conclusion that K8L will annihilate Core2 though, so either you've got an ES sample you're not telling us about , or you have the utmost faith in AMD to deliver as promised.

Would I be correct in assuming you are an AMD shareholder?

IIRC, you were the one who was proclaiming what speed Intel will be releasing chips at next year...
I certainly don't think either chip will annhiliate the other...my own opinion is that they will remain fairly close (5-10% difference) if AMD releases what they have shown on paper.

I am both an AMD and Intel shareholder (as well as Sun, HP, and Apple...).

 

[harpoon84]

No, I stated the speeds that VR-Zone projected for Intel. I never said it was set in concrete (NOBODY can say that at this point), just that I found them reasonable.

Anyhow, these sort of threads always go around in circles with everyone throwing about their own predictions and speculations...

 

[Viditor]

Originally posted by: harpoon84
No, I stated the speeds that VR-Zone projected for Intel. I never said it was set in concrete (NOBODY can say that at this point), just that I found them reasonable.

Anyhow, these sort of threads always go around in circles with everyone throwing about their own predictions and speculations...

Actually, what you said was:
"Intel is aiming for 3.5 - 4GHz on 45nm DC and 3.46 - 3.73GHz on 45nm QC according to the latest roadmap"

You did clarify in a later post:
"I'm mainly going off the VR-Zone site. They have a lot of 'updated' roadmaps on everything AMD and Intel"

The whole point I'm trying to make here (and I'm really trying not to be picking on you here, mate...) is that what VR-Zone is publishing are absolutely NOT roadmaps! They are publishing WAGs as if they are roadmaps...notice that they have absolutely no sources listed.
This is the difference between rumour and fact...
While I agree that not everything that either Intel or AMD publish comes true, at least you have someone to point the finger at. In addition, they must be much more cautious in supplying their facts as they could conceivably be sued by their respective shareholders if they supplied false info.

The other point I was trying to make is that, while everyone certainly has an opinion, it really would be better (and much more productive) if posters could mention WHY they have those opinions...

For example, I have an enourmous amount of respect for Duvie and what he posts...but that doesn't stop me from pointing out when I think he's incorrect in a post. That disagreement isn't because I am standing up for my "team", but because I think I owe it to him to voice my argument...BTW, the same holds true for your good self, mate.

Edit: BTW, I am most appreciative when you point out my mistakes as well...the only problem I have is when the comments are based on "fanboyism" et al. If you find a flaw in my arguments (and you're right that sometimes there can be no real resolution), then please point them out, but rest assured I do not post comments based on "loyalty to a team". If you scan through some of my posts, you'll note that I have lambasted AMD "fans" as well...

 

[myocardia]

Originally posted by: Viditor

Originally posted by: myocardia
I'm still waiting on you to come up with a link that explicitly quotes any AMD executive at all saying that an AM2 mobo will support any processor besides the currently available AM2's. BTW, I've been waiting for quite a few months for this link, it seems.

Sorry, I didn't realise you were waiting...
Glad to help though, it's in the same press release where they announced the AM2+ socket.
BTW, don't bother looking because they haven't announced that yet either...the source for all of the articles on Google is the same source that was cited for AM2+ in the first place, namely HKEPC.

If so, then why do you continually either outright say, or post links to people who have no idea, saying that these chips will be compatible with first generation AM2's? BTW, I feel the same as you about this:

Originally posted by: Viditor
For example, I have an enourmous amount of respect for Duvie and what he posts...but that doesn't stop me from pointing out when I think he's incorrect in a post. That disagreement isn't because I am standing up for my "team", but because I think I owe it to him to voice my argument...

 

[Viditor]

Originally posted by: myocardia

Originally posted by: Viditor

Originally posted by: myocardia
I'm still waiting on you to come up with a link that explicitly quotes any AMD executive at all saying that an AM2 mobo will support any processor besides the currently available AM2's. BTW, I've been waiting for quite a few months for this link, it seems.

Sorry, I didn't realise you were waiting...
Glad to help though, it's in the same press release where they announced the AM2+ socket.
BTW, don't bother looking because they haven't announced that yet either...the source for all of the articles on Google is the same source that was cited for AM2+ in the first place, namely HKEPC.

If so, then why do you continually either outright say, or post links to people who have no idea, saying that these chips will be compatible with first generation AM2's?

Fair enough, but I think you missed my point. If you accept that there is such a thing as AM2+, then you must accept that they will be compatible...mainly because the same people who "announced" them are saying that they will be compatible.
There has been a hint (which I have posted a few times) wherein Henri Richard stated in an interview that AMD's 4x4 would be compatible with quad core as well.

BTW, I feel the same as you about this:

Originally posted by: Viditor
For example, I have an enourmous amount of respect for Duvie and what he posts...but that doesn't stop me from pointing out when I think he's incorrect in a post. That disagreement isn't because I am standing up for my "team", but because I think I owe it to him to voice my argument...

Cheers, mate...and thanks for the discussion!

 

[DrMrLordX]

Originally posted by: Viditor

The thing is that Kentsfield won't be the competition, that would be Cloverton (remember that the first K8L Quad Cores will be Opterons...).

Hmm, you don't think AMD will have a K8L-based FX chip out at launch? I'm sure they'll have at least one for 4x4 users, though in that case, Cloverton would still be the logical competition.

Admittedly, I don't remember seeing one on the roadmaps per se, but . . .

 

[Viditor]

Originally posted by: DrMrLordX

Originally posted by: Viditor

The thing is that Kentsfield won't be the competition, that would be Cloverton (remember that the first K8L Quad Cores will be Opterons...).

Hmm, you don't think AMD will have a K8L-based FX chip out at launch? I'm sure they'll have at least one for 4x4 users, though in that case, Cloverton would still be the logical competition.

Admittedly, I don't remember seeing one on the roadmaps per se, but . . .

A good question, and the short answer is I just don't know...
They might have one for 4x4 when they release the Opteron Quad cores in June, or they might wait for the other desktops in Q3...
If I had to guess, I would agree with you that they will have the 4x4 FX chips early...but it's still a guess. The thing is that Q2 is the slowest season of the year for desktop and gamers, and they might want to wait until Q3 to capitalize on it. I guess we'll see sometime in the next few months...

 

[spikespiegal]

While were all getting in line waiting waiting for AMD to roll over and die, I need to bring up a quick history refresher. As I do recall, it wasn't but a year ago I was building dual core X2 and Opteron Servers that cost 1/2 as much as their Intel Netburst counterparts, generated a fraction the heat, and had 25-50% performance increases.

I guess memories are short.

 

[myocardia]

Originally posted by: spikespiegal
While were all getting in line waiting waiting for AMD to roll over and die, I need to bring up a quick history refresher. As I do recall, it wasn't but a year ago I was building dual core X2 and Opteron Servers that cost 1/2 as much as their Intel Netburst counterparts, generated a fraction the heat, and had 25-50% performance increases.

As long as that still applies today, then AMD has absolutely nothing to worry about.

 

[formulav8]

Looks like the Intel fannys are in denial. The truth of the matter is, IT WILL be a beast. But no one except AMD knows where its going to be exactly. At least according to specs, it will be VERY competitive with the Core2 in quite a few area's. Especially Integer apps. And if AMD can really, and I mean really make use of all that bandwith, it could really be a obsolute monster. But who knows for sure? They have a brand new arch. coming out in 2008 according to a interview. So they may not be putting all the tweaks in K8L that they could and even should to make sure they get their new core out in the next couple years. Just my opinion, obviously.


Jason

 

[Viditor]

Originally posted by: formulav8
Looks like the Intel fannys are in denial. The truth of the matter is, IT WILL be a beast. But no one except AMD knows where its going to be exactly. At least according to specs, it will be VERY competitive with the Core2 in quite a few area's. Especially Integer apps. And if AMD can really, and I mean really make use of all that bandwith, it could really be a obsolute monster. But who knows for sure? They have a brand new arch. coming out in 2008 according to a interview. So they may not be putting all the tweaks in K8L that they could and even should to make sure they get their new core out in the next couple years. Just my opinion, obviously.


Jason

Actually, that whole new arch has been brought forward...
What all of us were calling K8L (due in Q2 07) is actually going to be K10 according to this article. This does make a lot more sense as the number of changes they listed is far too large for just a revision...and remember that it's being called Rev B, not Rev H. In addition, K10 was originally designed to be a server chip first (which this will be)...I suspect that K8L will be a modified K8 design for mobile parts.

 

[gwag]

Performance can be meaningless to profits. I'd wager Intel makes more money selling graphics chips than Nvidia or AMD simpy because the sell the majority of chipsets for their CPUs. AMD is now heading towards that same path in a real way, and they will steal a lot of business from Nvidia and Via over the next year.
Selling more chips than you can make is a good thing they are still strong in the low and mid range computers and seems to me they have a better market presence than ever. Even if Intel makes faster better stuff they still have a lot of old stuff to dump in retail channel and in inventory, AMD is more streamlined with inventory.

 

[zsdersw]

Originally posted by: formulav8
The truth of the matter is, IT WILL be a beast.

That's a pretty matter-of-fact statement about a chip that we haven't seen yet. After this statement, you get more wishy-washy about it, which is fine because that's all we can do at this point, but the sentence of yours I quoted just doesn't square.

 

[coldpower27]

Originally posted by: Viditor

Originally posted by: formulav8
Looks like the Intel fannys are in denial. The truth of the matter is, IT WILL be a beast. But no one except AMD knows where its going to be exactly. At least according to specs, it will be VERY competitive with the Core2 in quite a few area's. Especially Integer apps. And if AMD can really, and I mean really make use of all that bandwith, it could really be a obsolute monster. But who knows for sure? They have a brand new arch. coming out in 2008 according to a interview. So they may not be putting all the tweaks in K8L that they could and even should to make sure they get their new core out in the next couple years. Just my opinion, obviously.


Jason

Actually, that whole new arch has been brought forward...
What all of us were calling K8L (due in Q2 07) is actually going to be K10 according to this article. This does make a lot more sense as the number of changes they listed is far too large for just a revision...and remember that it's being called Rev B, not Rev H. In addition, K10 was originally designed to be a server chip first (which this will be)...I suspect that K8L will be a modified K8 design for mobile parts.

I disagree what we have been describing as K8L is what AMD call's K10, which still looks to be a enhanced version of K8, when I mean enhanced I mean, that there are some considerable changes and not minor tweaks on the architecture front.

There is no change in the schedule, K8L, now referred to as K10 is slated for a Q2 2007 launch for the Server part Bareclona and the desktop SKU's are slated for release in the Q3 2007. Nothing has changed besides the fact that now we learn what AMD themselves are calling the architecture improvements they have down the pipeline for Mid 2007.

 

[coldpower27]

Originally posted by: spikespiegal
While were all getting in line waiting waiting for AMD to roll over and die, I need to bring up a quick history refresher. As I do recall, it wasn't but a year ago I was building dual core X2 and Opteron Servers that cost 1/2 as much as their Intel Netburst counterparts, generated a fraction the heat, and had 25-50% performance increases.

I guess memories are short.

A year ago AMD's desktop Dual Core line was still priced much more expensive then what Intel had to offer, it had better performance mind you but it was not cheap. It certainly was not 1/2 as much as their Intel counter parts. For the server yes I can agree as all Intel had was Paxville.

Regarding memories it seems from what I can gather here, people remember what they want to remember and not what actually happened. This happens with both sides and is not isolated to one group.

 

[coldpower27]

Originally posted by: DrMrLordX

Originally posted by: Viditor

The thing is that Kentsfield won't be the competition, that would be Cloverton (remember that the first K8L Quad Cores will be Opterons...).

Hmm, you don't think AMD will have a K8L-based FX chip out at launch? I'm sure they'll have at least one for 4x4 users, though in that case, Cloverton would still be the logical competition.

Admittedly, I don't remember seeing one on the roadmaps per se, but . . .

Considering Bareclona is slated for the DP and MP environments it's competition will be both Clovertown and Tigerton.

Since 4x4 is a LGA1207 platform, then it should be in the relam of possibility for AMD to put Barcelona cores on this platform a system with 2P 8C configuration. Whether they do this we will have to see, it certainly would be good PR getting out 8 core systems out on the desktop before Intel.

 

[myocardia]

Originally posted by: formulav8
Looks like the Intel fannys are in denial. The truth of the matter is, IT WILL be a beast.

Hmm, pot calling the kettle black, I see. The truth of the matter is it might be a beast. I will wait until both K8L and K10 are available, before making up my mind on the matter. Of course, that may just be because the last product I bought from Intel was my Pentium 233, in 1998.

 

[Viditor]

Originally posted by: coldpower27

I disagree what we have been describing as K8L is what AMD call's K10, which still looks to be a enhanced version of K8, when I mean enhanced I mean, that there are some considerable changes and not minor tweaks on the architecture front.

There is no change in the schedule, K8L, now referred to as K10 is slated for a Q2 2007 launch for the Server part Bareclona and the desktop SKU's are slated for release in the Q3 2007. Nothing has changed besides the fact that now we learn what AMD themselves are calling the architecture improvements they have down the pipeline for Mid 2007.

I guess it's personal perspective...you can also say that the C2D is an enhanced Piii (though I certainly wouldn't...). K10 was to be a "from the ground up" design, and while it's obviously strongly rooted in K8, I suppose that we shall have to wait for the samples to be released before we know how true that is...

 

[Viditor]

Originally posted by: coldpower27

Originally posted by: DrMrLordX

Originally posted by: Viditor

The thing is that Kentsfield won't be the competition, that would be Cloverton (remember that the first K8L Quad Cores will be Opterons...).

Hmm, you don't think AMD will have a K8L-based FX chip out at launch? I'm sure they'll have at least one for 4x4 users, though in that case, Cloverton would still be the logical competition.

Admittedly, I don't remember seeing one on the roadmaps per se, but . . .

Considering Bareclona is slated for the DP and MP environments it's competition will be both Clovertown and Tigerton.

Since 4x4 is a LGA1207 platform, then it should be in the relam of possibility for AMD to put Barcelona cores on this platform a system with 2P 8C configuration. Whether they do this we will have to see, it certainly would be good PR getting out 8 core systems out on the desktop before Intel.

Strangely, that's the one thing we know the most about...
Henri Richard talked about the quad core upgrade at 4x4's announcement.
Richard interview

While the platform will be limited to two dual-core processors, for a total of four cores, Richard promised an expansion for 2007. "What's really cool [about the technology] is that it's quad-core compatible, which means that sometimes next year you will be able to move from 4-core to eight-core."