AMD has a master plan that spells doom to Nvidia

[Skynet]

lol yeah I might be crazy.

Maybe. But we can always count on you to crap on anything AMD does all the while pretending you are impartial.

@Adored very interesting video I agree with a lot of what you are saying. I don't think Nvidia is in any serious trouble going forward but AMD's "Trojan Horse" strategy has all but guaranteed survival.

 

[razor1]

Just simple math and logic can show you Pascal has a lot of flexibility based on the current rumors, when it comes to clocks and power usage. Of course I know logic and math aren't your strong suites.

AMD is behind the q-ball when it comes to architecture and power usage and frequency, based on all rumors and what AMD has stated, I don't see very much coming from an architectural point of view for perf/watt (15-20%) at most. That puts it close to maxwell 2's range of perf/watt. Is this why they have only showed Hitman and Star War's battlefront to show off their perf/watt? Its possible, those two games AMD has a huge advantage with performance. And that will skew their perf/watt numbers. Added to that adding in the frame rate lock for smaller polaris in SWB , it skews the perf/watt numbers even more.

Has nV just sat around with their architecture? No they haven't, p100 if the numbers are real, and since nV is the one that gave the numbers most likely they are, they are getting 20% more from Pascal over Maxwell 2 just from base changes in the architecture. Added to that because the specs that nV released for p100 the shader clocks are much higher 40-50% higher, with more shader units and DP units. We know the DP units suck a lot of power and space in the silicon. We don't know if p100 is going to be released in the gaming space, if they are planning on a chip with no DP units, they save that space or use that space to put in SP units and still keep the clocks up at 1500mhz and that will drop the 300 watts to something much less.

That is what Big vega will be going up against. Once GP104 is released, AMD and we will have a very good idea of what Vega Big will be up against.

Also once Polaris is released we will have an excellent idea of what Vega will bring to the table from a power usage point of view.

 

[Skynet]

Just simple math and logic can show you Pascal has a lot of flexibility based on the current rumors, when it comes to clocks and power usage. Of course I know logic and math aren't your strong suites.

Logic and math based on rumours are you serious now?

 

[bittermann]

Logic and math based on rumours are you serious now?

Yes, pages and pages of made up facts from made up facts.

 

[razor1]

Logic and math based on rumours are you serious now?

You don't think AMD isn't behind nV when it comes to power and pushing the mhz? if you don't think they aren't behind, that is pretty bad logic. There is no rumor there. nV's architecture is just made for higher clocks and lower power usage. Straight facts and both of these aspects are not mutual to each other, its two different approaches that Maxwell 2 was designed for. AMD is so far behind in both of those right now, its not funny and its not an accident either. It wasn't the node that gave nV those advantages. Do you see Kepler or Fiji capable of doing what Maxwell 2 did with power usage and frequency? Both of those are on the SAME process. The only thing left is the architecture that is what gave it the advantage. And if you think AMD will be able to do the same, I can see them gaining some ground but not all of it. I think nV learned a lot of that doing tegra low power chips which translated directly to what they did with Maxwell 2. Knowledge like that doesn't come out of thin air. This type of knowledge AMD doesn't have access to as they weren't focused on it as there was no need for them.

These two advantages will come over to Pascal, both which AMD has to counter. I can see them doing one, not both. because again, it took nV how many attempts and how many years of Tegra development to get that knowledge. AMD isn't going to be able to do it in a year or two after they saw Maxwell 2. They can learn some of those approaches and incorporate them yes, not all of them though. And if you think they were thinking about them before that. I don't think they were since the need wasn't there. Its a completely different design philosophy. Remember what the architect for Zen stated about ARM vs. x86 design and trying to combine the two into one socket..... This is the same thing AMD's graphics division has to go through and more because nV has more than one generation of low power chip experience.

 

[bittermann]

We get it. AMD is bad, Nvidia is good. Its not like AMD is focusing on power vs performance efficiency this time around like those "factual" NV rumors you like to post. Nope, can't learn anything from the past.

http://wccftech.com/amd-unveils-polaris-11-10-gpu/

 

[razor1]

We get it. AMD is bad, Nvidia is good. Its not like AMD is focusing on power vs performance efficiency this time around like those "factual" NV rumors you like to post. Nope, can't learn anything from the past.

http://wccftech.com/amd-unveils-polaris-11-10-gpu/

Yet the 2.5 x increase in perf/watt is the same as the 60% drop in power usage the node gives!

What does that tell you about the architecture?

Wouldn't nV have the same advantage when they go down to 16nm? Yes they do!, So they also get the 2.5 x increase in perf/watt!

Everyone is focused on what AMD is saying, but no one is looking at what the node gives you right off the bat.....

And if 2.5x perf/watt increase is the best case AMD can get.......

 

[bittermann]

Yet the 2.5 x increase in perf/watt is the same as the 60% drop in power usage the node gives!

What does that tell you about the architecture?

Wouldn't nV have the same advantage when they go down to 16nm? Yes they do!, So they also get the 2.5 x increase in perf/watt!

Everyone is focused on what AMD is saying, but no one is looking at what the node gives you right off the bat.....

"This indicates that the Polaris video card is only using roughly 50W, making it twice as power efficient as the Nvidia Maxwell based GTX 950 graphics card. Which is a very significant power efficiency lead, mind thanks to 14nm FinFET technology and intelligently power efficient architectural design. Finally, the Polaris graphics chip tested was an early engineering sample and both hardware and driver optimizations are still on-going so additional power efficiency improvements can be expected".

It tells me you don't have all the facts yet. What does that tell you?

 

[razor1]

"This indicates that the Polaris video card is only using roughly 50W, making it twice as power efficient as the Nvidia Maxwell based GTX 950 graphics card. Which is a very significant power efficiency lead, mind thanks to 14nm FinFET technology and intelligently power efficient architectural design. Finally, the Polaris graphics chip tested was an early engineering sample and both hardware and driver optimizations are still on-going so additional power efficiency improvements can be expected".

It tells me you don't have all the facts yet. What does that tell you?

With frame lock on? Do you know without frame lock on 950 with the settings they used, gets over 90fps in that game? What is the Perf/watt advantage again? And please look it up, we talked about this enough at B3D, and came to the same conclusion. That preview didn't tell us Jack about Polaris. It could be anywhere from 0 times onward when compared to the 950.

And the 950 is not the most perf/watt chip in its family.....

Yeah AMD used the best case in that preview. That should tell you something.

 

[bittermann]

With frame lock on? Do you know without frame lock on 950 with the settings they used, gets over 90fps in that game? What is the Perf.watt advantage again?

And the 950 is not the most perf/watt chip in its family.....

Yeah AMD used the best case in that preview.

Please prove its strictly the node improvement. They stated the Maxwell 900 series, not just the 950, which is just the one in the comparison. Since nothing on that node is released yet that would be a pretty nifty feat.

So basically you don't believe AMD but you do Nvidia...got it.

 

[razor1]

Please prove its strictly the node improvement. They stated the Maxwell 900 series, not just the 950, which is just the one in the comparison. Since nothing on that node is released yet that would be a pretty nifty feat.

What you haven't seen AMD's press releases, Forbes talking about the node advantages, Samsung talking about it, and GF talking about it? They all did. Here is Samsung talking about its 14nm process, both

http://www.samsung.com/semiconductor/foundry/process-technology/14nm/

14LPE (Early edition) targets the early technology leaders and time-to-market customers such as mobile application SoCs to meet the latest mobile gadgets’ aggressive schedule and improved performance/power requirements. 14LPE is the first foundry process technology manufactured in the foundry industry with the successful volume ramp-up. 14LPE offers 40% faster performance; 60% less power consumption; and, 50% smaller chip area scaling as compared to its 28LPP process.

14LPP (Performance boosted edition) is the 2nd FinFET generation which the performance is enhanced up to 10%. 14LPP is the single platform for every application designs with the improved performance for computing/Network designs and the lowered power consumption for Mobile/Consumer designs. 14LPP will be the main process technology offering in 2016 and after.

the semi colon between 40% faster performance and 60% less power consumption denotes its either or. So there it is your 2.5times perf /watt change.... 60% less power consumption.

AMD did get that advantage in best case, but nV gets the same advantage in best case too as the 16nm process TSMC states 45% increase in performance or 70% drop in power consumption but I think they fudged the numbers a bit to make themselves look better. As 16nm process was built by Samsung to begin with.

 

[The Mac]

no, the semicolon does not indicate "or".

semicolons indicate lists of closely related ideas.

lists by their nature are inclusive.

if they were indicating what you are implying, they would have actually used a comma, and an actual "or" to differentiate from the and before the size.

that quote flat out says 40% faster, 60% less power, and 50% smaller.

not only is your math bonkers, your grammar stinks too.

lol

as a side note, the actual quote is grammatically incorrect anyway. it should read "14LPE offers 40% faster performance, 60% less power consumption, and 50% smaller chip area scaling as compared to its 28LPP process." Perhaps the joker who wrote it thought it would be funny to confuse a bunch of tech nerds. Or the copy editors in South Korea stink.

 

[badsykes]

Oh common guys ... can you stop this throwing the cat from one another...Patience for a couple of weeks and all things will be revealed ....

razor1: you know you cannot convince another one until they convince themselves ... They really want AMD to be in the front now... This may be a little illusion they want to have ... So let them

 

[razor1]

no, the semicolon does not indicate "or".

semicolons indicate lists of closely related ideas.

lists by their nature are inclusive.

if they were indicating what you are implying, they would have actually used a comma, and an actual "or" to differentiate from the and before the size.

that quote flat out says 40% faster, 60% less power, and 50% smaller.

not only is your math bonkers, your grammar stinks too.

lol

as a side note, the actual quote is grammatically incorrect anyway. it should read "14LPE offers 40% faster performance, 60% less power consumption, and 50% smaller chip area scaling as compared to its 28LPP process." Perhaps the joker who wrote it thought it would be funny to confuse a bunch of tech nerds. Or the copy editors in South Korea stink.

No a semicolon gives equal importance to both sides as it pauses the sentence. That is what grammatically correct is, I don't know where you learned your English but you are incorrect. When used that way that is what you get, that's why the next semicolon has and and next to it, because the 50% is also just as important but since the and is there, its with one of the other features inclusive!

My math is 100%, just like my grammar, you my friend, don't know **** about business, don't know **** about math, don't know **** about grammar and if you want to go back further don't know **** about programming (remember the whole DXGI debacle, yeah I haven't forgotten). As you seem to have always been on the short end of the dumb stick.

Just for you Mac.

http://www.colonsemicolon.com/

The most common use of the semicolon is to join together two clauses that could each be separate sentences — creating a longer sentence. For example:
John calls it football; Sam calls it soccer.
This could be written as two sentences without the semicolon; however, the relationship between the two clauses is made more clear through the use of a semicolon. The semicolon is often used to make the reader think about the relationship between the two clauses.
The semicolon is also commonly used to join two clauses, changing the sentence in combination with words like ‘therefore’, ‘however’ or ‘on the other hand’. The examples below illustrate this approach:

 

[razor1]

Oh common guys ... can you stop this throwing the cat from one another...Patience for a couple of weeks and all things will be revealed ....

razor1: you know you cannot convince another one until they convince themselves ... They really want AMD to be in the front now... This may be a little illusion they want to have ... So let them

Yes them being blind as a bat really helps them I can see.

 

[bittermann]

Yes them being blind as a bat really helps them I can see.

My only point is your pulling numbers out of your arse until we have actual reviews. Its easy to gloat on either side with nothing concrete to back it up. And yes, in a couple weeks we should find out, which is great news for those that have been waiting.

Finally, the Polaris graphics chip tested was an early engineering sample and both hardware and driver optimizations are still on-going so additional power efficiency improvements can be expected.

I'm hoping for the best from both vendors. That is a win for all gamers, not just you fan boys.

 

[The Mac]

No a semicolon gives equal importance to both sides as it pauses the sentence. That is what grammatically correct is, I don't know where you learned your English but you are incorrect. When used that way that is what you get, that's why the next semicolon has and and next to it, because the 50% is also just as important but since the and is there, its with one of the other features inclusive!

http://www.colonsemicolon.com/

you are batshit crazy.

there is absolutely nothing in your link that says a semicolon means "or". (Or any other technical writing book i have read)

If you are joining sentence fragments with a semicolon, each fragment must be explicit. Since none of the fragments in the qoute explicitly say they are non-inclusive, it its deems to be inclusive.

http://theoatmeal.com/comics/semicolon
https://writing.wisc.edu/Handbook/Semicolons.html
https://en.wikipedia.org/wiki/Semicolon

Im guessing you had a public education?

Propper grammer should NEVER use a semicolon when a comma would suffice. Each of those fragments ARE NOT separate ideas, therefore a semicolon is inappropriate

Like i said, the original quote is also grammatically incorrect.

My original statement stands.

Believe what you want, i could care less at this point as we are way OT again..

 

[Higgy10]

Enough grammar nazi **** plz. Ftr Mac is right, thats the way I learned English at least.

 

[razor1]

Coordination Relates "Equal" Ideas

Coordination gives equal emphasis to different ideas in a sentence. Coordinating conjunctions (and, but, for, nor, or, so, and yet) or a semicolon. (The semicolon is often used with a conjunctive adverb such as therefore, moreover, or however.)

Bernard brought John back to the World State and exhibited him like a circus freak.
John sought to live free of the World State, but could not escape from the intrusive interest of others.
Deckard is not an android; however, he is not fully human either.
​

http://web.calstatela.edu/faculty/jgarret/sentences.htm

I guess they don't teach you this in your college, but yet, other colleges are teaching it........

The reason they used the semicolon in the first place, is because they used another after with a conjunction of add

It was appropriately used, if the second semicolon and "and" wasn't there then it would have been improperly used, as this would have given two or more meanings.

 

[razor1]

My only point is your pulling numbers out of your arse until we have actual reviews. Its easy to gloat on either side with nothing concrete to back it up. And yes, in a couple weeks we should find out, which is great news for those that have been waiting.

Finally, the Polaris graphics chip tested was an early engineering sample and both hardware and driver optimizations are still on-going so additional power efficiency improvements can be expected.

I'm hoping for the best from both vendors. That is a win for all gamers, not just you fan boys.

Yet you have an issue when I say small Vega may or may not beat pascal gp 104 at the higher end of the spectrum but the outlook looks promising. What gives? Like to slide in there when you feel its ok and you can make it sound like it doesn't support what you think 100%, and when asked for clarification, you just dismiss everything that has been well documented by the Fabs about their process and what AMD stated about their max perf/watt increase because it doesn't fit in with hopes and figments of your imagination of AMD improving their perf/watt performance through architecture? I can see where this is going...... You believe one side of AMD marketing but you don't believe the marketing from the fabs?

Mind you I'm not saying they aren't going to get any increased throughput they are going to get some just not as much as 2.5 perf/watt, more like a 20% increase. Which 20% is a large number and a similar number to nV got from Keplar to Maxwell 2.


http://www.pcworld.com/article/3020...will-survive-and-yes-even-thrive-in-2016.html

There’s no question that process technology has been a big knob to turn, in the performance and power spectrum,” Su added. “That’s why we went to FinFET, that’s why we went to FinFET relatively early, because we saw the power benefits.”

http://www.decryptedtech.com/news/g...nces-14nm-process-in-partnetship-with-samsung

Now, before everyone gets all excited, there are some limitations on this new node. There is no cost savings by moving to this yet. While the major process is 14nm is would appear that they still use 20nm interconnects. It is an interesting mix of technologies, but you still should get a decent performance boost for the same power usage.