AMD reportedly finishes testing Zen CPU, claims they met expectations

[dividebyzero]

I really want to see (and hope ) that software developers finally get it together and widely start writing programs that take advantage of the cores they now have at their disposal.

Yeah, good luck with that. Software houses don't generally throw tuned code out to the masses unless it can attract a return - either directly because job time-to-completion is a paramount concern, or because the software aids in marketing (I.e. as a widely used benchmark). For the most part, adding cores to a consumer part for consumer workloads is a brute force approach to bad coding and mass marketing - as I suspect you already know...and will remain so for the foreseeable future.

 

[Puiu]

Yeah, good luck with that. Software houses don't generally throw tuned code out to the masses unless it can attract a return - either directly because job time-to-completion is a paramount concern, or because the software aids in marketing (I.e. as a widely used benchmark). For the most part, adding cores to a consumer part for consumer workloads is a brute force approach to bad coding and mass marketing - as I suspect you already know...and will remain so for the foreseeable future.

don't worry, new software in general is made to take advantage of multicore CPUs because they use new APIs and libraries that have this feature by default. it's just hard to update older software.

 

[dividebyzero]

don't worry, new software in general is made to take advantage of multicore CPUs because they use new APIs and libraries that have this feature by default. it's just hard to update older software.

Oh, I know there are an increasing number of software packages that can take advantage of multi threaded processors, it's just that consumer (as opposed to WS/server) software- while it might use multiple threads, it might not be using them particularly effectively. There is also a large base of supposedly multi-core aware software that isn't particularly well optimized for SMT - indeed, there are plenty of instances where the code is so well unoptimized that multi-threading actually yields worse results.

 

[red1776]

Yeah, good luck with that. Software houses don't generally throw tuned code out to the masses unless it can attract a return - either directly because job time-to-completion is a paramount concern, or because the software aids in marketing (I.e. as a widely used benchmark). For the most part, adding cores to a consumer part for consumer workloads is a brute force approach to bad coding and mass marketing - as I suspect you already know...and will remain so for the foreseeable future.

ah, a girl can dream. I think it will get better, however in an Adobe sort of way.

 

[rpohl67]

Too many whiners out there. Why don’t you wait till the product comes out and then comment? I am a huge AMD fan because I fill that when I buy a processor I am buying the processor not the name. This is the reason AMD sells their processors cheaper your not paying for the name. Even when AMD was leading over Intel their processors were cheaper and Intel had to lower their prices just to compete Sounds Familiar right.

 

[dividebyzero]

Too many whiners out there. Why don’t you wait till the product comes out and then comment? I am a huge AMD fan because I fill that when I buy a processor I am buying the processor not the name. This is the reason AMD sells their processors cheaper your not paying for the name. Even when AMD was leading over Intel their processors were cheaper and Intel had to lower their prices just to compete Sounds Familiar right.

Maybe you should read through the thread. Another poster made the same inference, which is actually untrue as a quick perusal of the Intel and AMD product lists will show for the times when AMD edged Intel. Pricing for both sides equated roughly equal based on clockspeed and market segment - with some occasional outliers due to high demand by OEMs or low yields.
AMD actually raised pricing to Intel's levels once they had processors of an equal or greater performance.

 

[rpohl67]

Maybe you should read through the thread. Another poster made the same inference, which is actually untrue as a quick perusal of the Intel and AMD product lists will show for the times when AMD edged Intel. Pricing for both sides equated roughly equal based on clockspeed and market segment - with some occasional outliers due to high demand by OEMs or low yields.
AMD actually raised pricing to Intel's levels once they had processors of an equal or greater performance.

I don’t need to read through the threads because I am giving my opinion not asking for help. Pricing for both sides was not equal. Intel needed to lower their prices several times. AMD processors were selling so good Intel had to lower prices just to keep up. At least that is what I read in computer magazines or maybe their telling us lies Oh the conspiracies.

 

[dividebyzero]

I don’t need to read through the threads because I am giving my opinion not asking for help. Pricing for both sides was not equal. Intel needed to lower their prices several times. AMD processors were selling so good Intel had to lower prices just to keep up. At least that is what I read in computer magazines or maybe their telling us lies Oh the conspiracies.

If you are reading computer magazines then the pricing aggregating service many run (or did run at the time) would show price fluctuations and (then) current pricing. Fluctuations from month to month (and week to week on occasion) varied wildly compared to the staid pricing currently available - it all depended upon yields and launch cadence. Early 2000 you would see parity, but AMD dropped prices late in the year as Thunderbird came on stream (and earlier K7 / K75's were heavily price cut as well) and Intel was caught between older Coppermine and the Tualatin PIII's release which was still months away. Anyhow, you said you have the magazines, so pricing comparisons should be straightforward.

Spoiler

 

[rpohl67]

If you are reading computer magazines then the pricing aggretating service many run (or did run at the time) would show price fluctuations and (then) current pricing. Fluctuations from month to month (and week to week on occasion) varied wildly compared to the staid pricing currently available - it all depended upon yields and launch cadence. Early 2000 you would see parity, but AMD dropped prices late in the year as Thunderbird came on stream (and earlier K7 / K75's were heavily price cut as well) and Intel was caught between older Coppermine and the Tualatin PIII's release which was still months away. Anyhow, you said you have the magazines, so pricing comparisons should be straightforward.

Spoiler

I am not going to argue prices that aren’t important from days long ago. Let us just see how the new processors performs before we start criticizing them. I have had a lot of AMD processors that run games maxed on settings and they run flawlessly. Just Saying. Thank you

 

[Axiarus]

Yeah once conroe hit bam! that was it. although phenom II was a great comeback and the last good amd proc.

Phenom II was still behind though.

 

[Geforcepat]

Phenom II was still behind though.

Yeah I know. but it was still better for it's time than Faildozer

 

[maxxxenem1]

AMD sucks Intel and Nividia are king. Period! Amd is always and has always been behind the curve.

 

[red1776]

AMD sucks Intel and Nividia are king. Period! Amd is always and has always been behind the curve.

I just think Argumentum ad lapidem such as this are just grand.

 

[dividebyzero]

I just think Argumentum ad lapidem such as this are just grand.

I think someone took a wrong turn at the WCCFtech forum exit ramp.

 

[red1776]

I think someone took a wrong turn at the WCCFtech forum exit ramp.

hehehe, or the FUD off ramp.