Deneb just around the corner?

[charged3800z24]

I think Penryn across the board was about 5 to 8%.

OK, well add 4mb cache and we should see a more apps see more improvement. maybe 12 - 18 in some case... I am to lazy to figure this out...but my Phenom @ 3.5ghz had a Pi 1m run in 22.5ish area with 2400mhz NB and 240 HT ref clock and a little better timings on the ram. And the Deneb C0 stepping a while back hit just over 22secs in Pi 1m with a 3.2ghz clock and a lot less NB etc. In the particular bench mark it show it does have improvements and it wasn't tweaked like mine. I should probably do a same clock comparison to it and figure the actual % improvement. I know, Pi 1m is not a favored bench to some, but in same compay archtecture comparison, it can show some hint of performane gain, right?

[roofsniper]

I am casting my doubts for two reasons, and if you would care to address those reasons that would be good, and if you don't want to, that's okay too.

1. In all previous shrink + Cache increases by either side in the last 12 to 15 years at least, we haven't even come close to seeing a 15 to 20% IPC increase, what is going to make Deneb so different?

2. AMD have little to no credibility in respect of projected CPU performances for at least 18 months now.

wow how many times do we have to say its not just a shrink + cache change? deneb is different because it is more than just that. there have been modifications to the cpu that are made that should of been made with agena and are new modifications that they have made based on the last cpu. at the time amd started making the first phenom they were already ahead of intel with their X2 cpus. amd was obviously careless and were determined to make a native quad core. it took them a lot of time to do so and when they finally got it done it was a mess but they were so far behind already that they just had to release the chip because of the huge delay. deneb is supposed to be the first real k10. they have had a lot more time to do what they needed to do and they have done it now. and this time there is no tlb bug. they can actually release the chip without knowing it has a handful of problems. i have always considered the first phenoms to be a prototype. even the day they were released and i still believe them to be the same today.

ok and your second statement is just asking for trouble. you obviously don't know a thing about amd. and they have lived up to the performance they have told people. really i have no idea why you have come in here and posted. you have no facts. you refuse to believe anyone's opinion and deny all facts presented to you. it sounds like you are just trolling and trying to start something because you will not be satisfied until everyone in here says amd sucks. you have nothing constructive to say.

[Titan7171]

ok so you believe that 15-20% is excessive. amd has stated that thats what it is and i have spoken with amd employees about it and they say the same. because you have no proof that 15-20% is excessive then i don't believe you can say that others are wrong when they claim this. we all accept that you believe that the ipc will not increase by 15-20% but what we don't understand is how u can say we are all wrong without having any proof stating otherwise when everyone else here is getting all their numbers and performance predictions from facts and what amd has stated.

This dude needs attention why do you even bother responding he doesnt believe its going to be 10-15% increase....... great we heard him why does he keep sayng the same thing over

So what if this guy doesnt believe what AMD says or that he thinks we are wrong who really cares?......hope he leaves soon

OMG how do you ignore a user?

[Chad Boga]

wow how many times do we have to say its not just a shrink + cache change?

So AMD is going to tidy up some things as well, but that is often if not always done when shrinks occur.

I believe you are placing too much emphasis on this.

ok and your second statement is just asking for trouble. you obviously don't know a thing about amd. and they have lived up to the performance they have told people.

You mean like their claim that Barcelona would be 40% faster than Core 2, when that only applied to bandwidth?

really i have no idea why you have come in here and posted. you have no facts. you refuse to believe anyone's opinion and deny all facts presented to you.

What are the actual facts presented to me that I am refusing to believe?

Why is it fine for you and others to refuse to believe my opinion, but not for me to disagree with other people's opinion?

One hell of a double standard you have there.

[roofsniper]

we all accept that you believe that the ipc will not increase by 15-20% but what we don't understand is how u can say we are all wrong without having any proof stating otherwise when everyone else here is getting all their numbers and performance predictions from facts and what amd has stated.

if you read this post then you will see that i am accepting your opinion but wondering why you are denying everyone else's. this is the last time i will respond to one of your posts as you obviously do not care what others think, are trolling, aren't using common sense, have no knowledge of the topic.

[Chad Boga]

if you read this post then you will see that i am accepting your opinion but wondering why you are denying everyone else's. .

How am I denying everyone else's opinion?

What would you have me do differently when I think Deneb will come in around 10% IPC improvements and others think 15 to 20%?

have no knowledge of the topic

What makes you think I have no knowledge of the topic?

Do you not think previous shrinks offer any sort of valid guide as to what we can expect for Deneb?

[Aussie FX]

If Deneb comes with a significant IPC boost, great, everybody wins.

But I remember all the talk about how great Thunderbird was going to be too back in the day and it was what, 4 to 5% improvement.

So colour me doubtful about any statements emanating from AMD PR, I wanna see the benchmarks.

Well go and look at them then, everyone else here has seen them, that's why we know about the improvements.

Google is your friend. Hint; try hardspell and coolaler.

It would also help if you tried to educate yourself a little.

[Chad Boga]

Well go and look at them then, everyone else here has seen them, that's why we know about the improvements.

Google is your friend. Hint; try hardspell and coolaler.

Official benchmarks.

[Aussie FX]

Official benchmarks.


You first.


It would also help if you tried to educate yourself a lot.

Oh come on Chad, people are trying to be nice but you're coming across now as being childish.
In the long run everyone will just ignore you. What if one day you need some help? I did just the other day with a flashing problem and I got helpful replies almost instantly.

It pays to be friendly and enjoy each other around here, starting silly arguments will get you nowhere.
Others here have attempted to steer you just as i'm attempting to now, but you will only get so many chances.

[Chad Boga]

Oh come on Chad, people are trying to be nice but you're coming across now as being childish.
In the long run everyone will just ignore you. What if one day you need some help? I did just the other day with a flashing problem and I got helpful replies almost instantly.

It pays to be friendly and enjoy each other around here, starting silly arguments will get you nowhere.
Others here have attempted to steer you just as i'm attempting to now, but you will only get so many chances.

I really can't see how your previous post was being nice.

[Aussie FX]

I really can't see how your previous post was being nice.

Fair enough, I shouldn't have said it and I apologise.
I will remove the offending line.

[Nedjo]

gee I'm of for one day and this thread explodes on 6 pages thanks to Chad Boga's rambling

In my own book the biggest thing that I'm looking for with arrival of Deneb is significant boost in games that I'm (re)playing most often... and yes improved O'Cing capabilities will come quite handy

In light of Nehalems price and game performance I think that Deneb will become the most reasonable choice of gamers looking for balanced systems that have sufficient power in GPU and CPU part of the system.

[Chad Boga]

gee I'm of for one day and this thread explodes on 6 pages thanks to Chad Boga's rambling

In my own book the biggest thing that I'm looking for with arrival of Deneb is significant boost in games that I'm (re)playing most often... and yes improved O'Cing capabilities will come quite handy

In light of Nehalems price and game performance I think that Deneb will become the most reasonable choice of gamers looking for balanced systems that have sufficient power in GPU and CPU part of the system.

If AMD are able to fix the clock rate issues with Deneb, and there certainly seems to be some strong indication that they have, then I wonder if the Battle Royale will be between Deneb and E8500?

[YukonTrooper]

In my own book the biggest thing that I'm looking for with arrival of Deneb is significant boost in games that I'm (re)playing most often... and yes improved O'Cing capabilities will come quite handy

That's what I'm looking for. A dual-core that is ~15% better in games than the current C2D's, sold at a competitive price. Of course, it will have to overclock just as well as the C2D's, so that the ~15% improvement can stay as a ~15% improvement.

[soundood]

guys this guy has been saying the same thing for 4 pages now

only reason i can see what he sais is in your quotes, please ignore him, please

This Thread has now been escalated to a higher Risk



for use in future when an Intel Fanboy invades the forum,

[Aussie FX]

That's the killer for AMD.
They appeared to have dropped the ball with dualcore for some reason.
How many are there that are based around K10, is it 2? Then you can't get them anyway.
Tricore seems to be AMD's new dualcore but they're only a way of selling otherwise dud quads.

[Daveburt714]

soundood: You and I finally agree on something!!

Who woulda thought...

[Chad Boga]

Soundud,
I love your warning system. LOL!!!!

[Aussie FX]

You mean like their claim that Barcelona would be 40% faster than Core 2, when that only applied to bandwidth?

You are twisting the facts here.
Barcelona is actually 40% faster than the equivalent clocked Xeon in Specfp_rate which is exactly what AMD said in the first place.
This claim of 40% over C2D is completely false and the rumour came from the web with people having no understanding of what they are talkibg about. I hope that clears it up for you , so you to can now stop posting that falsehood.

[Doctor Pepper]

The IPC is one thing, congestion in some ”sub system” is another. I truly believe that AMD have taken the opportunity to optimize cache and memory handling.

Better IPC can give a boost in certain areas (computational intensive), while yielding less in other (heavy data shuffling). I believe AMD have addressed both problem in Deneb.

In other words, even if IPC gets the 15-20% boost (which for certain IS possible), if the rest of the “chain” are unable to handle these are numbers we will only see in synthetic benchmarks *coughIntelcough* and not in our everyday computer experience…

I would like to know if also the AM2+ Deneb are going to have higher NB/HT frequencies, or if this is only for the HT3.1 ready motherboards? In theory the HT2 allows for up to 2600 MHz. I can feel, on my computer, how much better everything flows with NB/HT at 2440 MHz compared to 1800/2000 MHz, so I am certain that this will be an important point!

So, my uneducated guess, supported by some (dubious) benchmarks shown on internet is this:

1. Bigger cache

2. Optimized cache handling

3. (hopefully) Some SSE and possibly INT optimizations

4. Faster NB/HT frequencies

5. All this at lower power draw…

With most of these whishes fulfilled, there should be no problem reaching a clock for clock boost of 12-15% on an average. Weighing in substantial headroom for higher clocks, AMD have in my book a CPU that is equally fast, or faster , than Intels offering for real life computer experience. Maybe not for all synthetic SSE4 optimized benchmarks, but that is another story…



And Chad Boga, I understand your standpoint and you bring up some valid questions. However when a discussion morphs into a “meta discussion” (a discussion about the discussion…) everyone have to try to let that stop after the first round because all that is left after that is name calling (which leads to nothing good…)

[qurious63ss]

You are twisting the facts here.
Barcelona is actually 40% faster than the equivalent clocked Xeon in Specfp_rate which is exactly what AMD said in the first place.
This claim of 40% over C2D is completely false and the rumour came from the web with people having no understanding of what they are talkibg about. I hope that clears it up for you , so you to can now stop posting that falsehood.

You have made some good points but on this one I will have to say you're wrong. This was a quote from Randy Allen back in 2007.


"We expect across a wide variety of workloads for Barcelona to outperform Clovertown by 40 percent," Allen said.

He used this broad term that got everyone really excited when in reallity barcelona was 40% faster but only in specfp_rate.

http://news.cnet.com/AMD-Go-to-Barce...3-6152645.html

http://www.dailytech.com/article.aspx?newsid=5863

http://www.theregister.co.uk/2007/06...arcelona_2ghz/

oh and of course: http://sharikou.blogspot.com/2007/04...e-running.html

the horses mouth: http://www.youtube.com/watch?v=G_n3wvsfq4Y

BTW I don't think Mr. Allen works for AMD anymore, I wonder why...

[Chad Boga]

And Chad Boga, I understand your standpoint and you bring up some valid questions. However when a discussion morphs into a “meta discussion” (a discussion about the discussion…) everyone have to try to let that stop after the first round because all that is left after that is name calling (which leads to nothing good…)

This is probably a difficult balancing act because quite clearly in this thread there have been people determined to shut down anything other than complimentary talk about AMD, so they would always turn things into a meta discussion due to their unreasonableness.

qurious63ss

Thank you very much for saving me the leg work.

[Oliverda]

You have made some good points but on this one I will have to say you're wrong. This was a quote from Randy Allen back in 2007.


"We expect across a wide variety of workloads for Barcelona to outperform Clovertown by 40 percent," Allen said.

He used this broad term that got everyone really excited when in reallity barcelona was 40% faster but only in specfp_rate.

http://news.cnet.com/AMD-Go-to-Barce...3-6152645.html

http://www.dailytech.com/article.aspx?newsid=5863

http://www.theregister.co.uk/2007/06...arcelona_2ghz/

oh and of course: http://sharikou.blogspot.com/2007/04...e-running.html

the horses mouth: http://www.youtube.com/watch?v=G_n3wvsfq4Y

BTW I don't think Mr. Allen works for AMD anymore, I wonder why...

They did not enable all features in K10 which were originally planned. IMHO the Shanghai/Deneb will be the real K10.

Chad Boga: When are you going to get back to your home section? Your stupid and empty posts are going to be so boring for everybody.

[Sentential]

So, my uneducated guess, supported by some (dubious) benchmarks shown on internet is this:

1. Bigger cache

2. Optimized cache handling

3. (hopefully) Some SSE and possibly INT optimizations

4. Faster NB/HT frequencies

5. All this at lower power draw…

I know that im going to butcher his name but Coolaler has made several posts on his website showing C1 Deneb samples. So far it appears that C0 had the standard 1800/2000 HTT whereas the retail C1 stepping's HT was 2600 by default

C0 was getting iirc rougly 3.6 but he had a screenie of a C1 Deneb @ 4.3 on a scythe orochai heatsink (air) on MSI's 790GX (SB750)

Most importantly was that the SuperPI times he showed with the C1 Deneb were on-par with Yorkfield chips of the same speed hence the hope we have for Deneb. Shortly after this made its way to several sites on the net:

http://www.nordichardware.com/news,8195.html

[Chad Boga]

Chad Boga: When are you going to get back to your home section? Your stupid and empty posts are going to be so boring for everybody.

I don't have a home section and I suggest you stop carrying on like AMD is your wife, it is really sad and immature.