Looks like a moot point now. I suspect you're probably correct, but I think it would have been worth looking at how much of a difference it makes in both directions
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
*** AMD "Zen" thread (inc AM4/APU discussion) ***
- Thread starter Boomstick777
- Start date
Looks like a moot point now. I suspect you're probably correct, but I think it would have been worth looking at how much of a difference it makes in both directions
Yeah, but she failed miserably and now knows that unless she wants Zen to be stillborn and for Intel to dominate the next 5 years like they did the last 5 (something AMD cannot afford) then she will have to face reality.
Soldato
3.6 base 4turbo. Not bad as long as we can overclock to 4700+ 
and costs less than 6900k.
Also someone above said something about 10nm...
AMD goes straight to 7nm end of 2018 for all products. GloFo already setting up the manufacturing process for mass production, ignoring 10nm as waste of time and resources.
and costs less than 6900k.Also someone above said something about 10nm...
AMD goes straight to 7nm end of 2018 for all products. GloFo already setting up the manufacturing process for mass production, ignoring 10nm as waste of time and resources.
Last edited:
3.6 base 4turbo. Not bad as long as we can overclock to 4700+
and is fast as you hope it is lol.
what does it matter what it oc to if it isnt as fast
Soldato
and is fast as you hope it is lol.
what does it matter what it oc to if it isnt as fast
It looks as fast as the BE clock for clock. Which is fast enough IPC compared to upgrade from 6700k.
Caporegime
based on ? what ....
a blender benchmark that was a mess and not clear which version they used
there are 100s of real benchmarks.not one was done.so it doesnt matter if it does 5ghz on air if its still slower.
what a load of old crap......
It was very clear what version they used, it was 1.77, you could even see the version number in the shell of Blender as the benchmark was running.
was there debate or things unclear about the blender benchmark ? yes/no ?
= yes.
if the cpu was faster or better. they would have done numerous benchmarks showing so.
they just shown a benchmark which if you want to say nothing was this or that fair enough but it favoured them in a way.so thats why you seen it.its that simple.
we will see soon.i think in odd benchmark the top one maybe close to some intel chips odd benchmarks but slower in gaming and everything else. slightly cheaper than intel.
intel will know how fast they are.they would have adjusted prices heavily if they were beating intel by a margin.to get rid of stock.they havent.it speaks volumes.
= yes.
if the cpu was faster or better. they would have done numerous benchmarks showing so.
they just shown a benchmark which if you want to say nothing was this or that fair enough but it favoured them in a way.so thats why you seen it.its that simple.
we will see soon.i think in odd benchmark the top one maybe close to some intel chips odd benchmarks but slower in gaming and everything else. slightly cheaper than intel.
intel will know how fast they are.they would have adjusted prices heavily if they were beating intel by a margin.to get rid of stock.they havent.it speaks volumes.
Caporegime
There was never much of a debate, it was always know to be 1.77, the only confusion there was is how many cycles AMD ran because the 200 initially cited didn't match their results, AMD clarified it was 150, that then matched.
Edit: AMD didn't actually cite anything, the default is 200, AMD set it to 150 but didn't say so when they released the file.
That was it, panic over.
Edit: AMD didn't actually cite anything, the default is 200, AMD set it to 150 but didn't say so when they released the file.
That was it, panic over.
Last edited:
Soldato
They then advised people to download 1.78a. Considering performance differs from version to version, that was a clerical level error on their part at least...
Caporegime
Well 8C at 3.6GHZ to 4.0GHZ specially in light of the AMD demo and the CPC results is a big jump in clockspeed.
That sample was running at 3.1GHZ to 3.3GHZ,so that means the 8C/16T should be able to match or exceed the Core i7 6900K in non-gaming benchmarks on average.
Also,even the gaming benchmarks run,had 4 of the six games being very lightly threaded,meaning the CPU was probably stuck at 3.3GHZ,so a 10% to 20% clockspeed increase should easily make it match or get past a Core i7 6900K and close to a Core i7 4790K and Core i7 6700K.
If AMD can get upto 4GHZ with an 8C/16T SKU,it also means that a 4C/8T has the possibility of starting at a higher base clockspeed too,even if they overvolt it a bit,ie,decrease performance/watt.
That sample was running at 3.1GHZ to 3.3GHZ,so that means the 8C/16T should be able to match or exceed the Core i7 6900K in non-gaming benchmarks on average.
Also,even the gaming benchmarks run,had 4 of the six games being very lightly threaded,meaning the CPU was probably stuck at 3.3GHZ,so a 10% to 20% clockspeed increase should easily make it match or get past a Core i7 6900K and close to a Core i7 4790K and Core i7 6700K.
If AMD can get upto 4GHZ with an 8C/16T SKU,it also means that a 4C/8T has the possibility of starting at a higher base clockspeed too,even if they overvolt it a bit,ie,decrease performance/watt.
Caporegime
Thats really blurry, i can't read what Clock speed they were running at, looks like the 4790K was overclocked? ^^^^^
I don't remember them doing that, can you point me to where they did this? its new to me and i just don't see why they would advise using a different version to the one they said and was known they used, it make no sense, i'd like to see it.
I don't remember them doing that, can you point me to where they did this? its new to me and i just don't see why they would advise using a different version to the one they said and was known they used, it make no sense, i'd like to see it.
Last edited:
Soldato
Probably because mistakes are made. It was in the small print on the slides used. Hence why all the media sites flocked to tell people to download that particular build. It's of no consequence now anyway, but doesn't change the fact.
Caporegime
Its all semantics,we already got some leaked results from CPC,who have a long history of leaking stuff months before it was released and have been accurate in the past. CPC actually used Blender and POV Ray in their test set-up,and their 8C/16T Ryzen sample was running at a lower clockspeed than the one AMD was using.
An 8C/16T CPU at 3.6GHZ to 4.0GHZ is pretty good going if you ask me.
Its much higher than I expected they would get an 8C/16T CPU - I was expecting something like a 3.4GHZ to 3.8GHZ range for the 8C/16T and slightly higher base clockspeeds for the salvaged parts.
It means if AMD can get 4C/8T SKUs to 4.0/4.2GHZ max Turbo,and say a base clockspeed of 3.8GHZ,AMD should get close enough to a Core i7 6700K/7700K which should do the job for most people especially if it is priced below the ripoff price for Intel 4C/8T CPUs.
Soldato
Yes, yes exactly. Nothing at all matters. Launch should be March by the way. Not sure if it's been posted by anyone here yet.
4Ghz on air is achievable, wouldn't expect much beyond that.
4Ghz on air is achievable, wouldn't expect much beyond that.
Caporegime
Right so completely unrelated to the CES demo.
Its easy to see how people can get confused if straw clutching.... but they really shouldn't, people should just stick to the facts as they are at the time, not mix and mash irrelevant crap from unrelated crap, seriously WTF?
Caporegime
CAT do you have the original image that screen-shot came from? looks like it was twice the size and shrunk, some important bits on it are unreadable.
Soldato
You really don't need to try and drown this one out, it's not that big of a deal. The fact remains that AMD did in fact list the incorrect version on the slides posted. That's tangible, something that actually happened. But we've seen in the past how that won't stop you
Caporegime
You really don't need to try and drown this one out, it's not that big of a deal. The fact remains that AMD did in fact list the incorrect version on the slides posted. That's tangible, something that actually happened. But we've seen in the past how that won't stop you
Its no deal at all, you brought it up despite it not being anything that matters. ^^^^ its not even related.