vm'N

You could shrink down the images to say, a thumbnail size, then do the comparison, using perceptual hashing.
More info here: http://dsp.stackexchange.com/questions/5995/what-algorithm-does-google-use-for-its-search-by-image-site

The 4790k is noit considered a haswell-e processor. It is considered a haswell refresh processor. Haswell-e is LGA2011 with 6-8 cores.
wikis page have a nice table to see what SKU goes under what segments:
https://en.wikipedia.org/wiki/Haswell_(microarchitecture)

Should they apply to more marketing to hype you up? Or wait until they have something?
These things take time, and there certainly are dry times, where much doesn't happend.

True, but i think it's safe to say it will be better than what they have now, hopefully a lot better

The 4790k is noit considered a haswell-e processor. It is considered a haswell refresh processor. Haswell-e is LGA2011 with 6-8 cores.
wikis page have a nice table to see what SKU goes under what segments:
https://en.wikipedia.org/wiki/Haswell_(microarchitecture)

i think that kitguru is just butthurt, cause he didnt get a furyX review sample lol.
 
#shitguru

The only thing I can take from all these acquisition rumours, is that a lot of companies are interested in AMDs portfolio rather than AMDs future.
To be honest, AMD have been kind of quite in regards to plans past a half a decade.

Who will be the next company to consider buying amd? Facebook, google, amazon?

You don't have to place it in a excel spreadsheet do you?
Else you can just format it right, and give it an .csv extention.
Then you can easily import it to any spreadsheet software.

The desktop one is better. The app is soo buggy and slow.

Okay. When your processor is doing a ton of work, and there will be a workload, like adding +18 to x amount of data-elements. Instead of going over it serialized (one after another), you load all the elements into your SIMD cluster, which then will do a single operation to all the data-elements in parallel.
 
Increased width means you can have more data-elements.
Gather-scatter is used to simplify the loading / storing of the elements for vector code.

AVX3 or AVX-512 is Intels upcomming AVX extention. Increase the width of the previous AVX extention (AVX2 or AVX-256) and introduce some better gather-scatter instructions IIRC.
 
If it means anything to you? No, not all all.

Okay. When your processor is doing a ton of work, and there will be a workload, like adding +18 to x amount of data-elements. Instead of going over it serialized (one after another), you load all the elements into your SIMD cluster, which then will do a single operation to all the data-elements in parallel.
 
Increased width means you can have more data-elements.
Gather-scatter is used to simplify the loading / storing of the elements for vector code.

AVX3 or AVX-512 is Intels upcomming AVX extention. Increase the width of the previous AVX extention (AVX2 or AVX-256) and introduce some better gather-scatter instructions IIRC.
 
If it means anything to you? No, not all all.

AVX3 or AVX-512 is Intels upcomming AVX extention. Increase the width of the previous AVX extention (AVX2 or AVX-256) and introduce some better gather-scatter instructions IIRC.
 
If it means anything to you? No, not all all.

Okay. When your processor is doing a ton of work, and there will be a workload, like adding +18 to x amount of data-elements. Instead of going over it serialized (one after another), you load all the elements into your SIMD cluster, which then will do a single operation to all the data-elements in parallel.
 
Increased width means you can have more data-elements.
Gather-scatter is used to simplify the loading / storing of the elements for vector code.

AVX3 or AVX-512 is Intels upcomming AVX extention. Increase the width of the previous AVX extention (AVX2 or AVX-256) and introduce some better gather-scatter instructions IIRC.
 
If it means anything to you? No, not all all.

AVX3 or AVX-512 is Intels upcomming AVX extention. Increase the width of the previous AVX extention (AVX2 or AVX-256) and introduce some better gather-scatter instructions IIRC.
 
If it means anything to you? No, not all all.

The court is not an agency. If the data is required for a fair trial i dont know why skype are being douchbags about it.

You do realize why AMDs workstation GPU is doing badly? Even through they keep tanking the price? AMD most certainly need CUDA. Without it, AMDs solution is no solution at all for many customers.
AMD could have tanked the price to 1/10th, and nvidia would still hold a majority of the marketshare.

You do realize why AMDs workstation GPU is doing badly? Even through they keep tanking the price? AMD most certainly need CUDA. Without it, AMDs solution is no solution at all for many customers.
AMD could have tanked the price to 1/10th, and nvidia would still hold a majority of the marketshare.

You do realize why AMDs workstation GPU is doing badly? Even through they keep tanking the price? AMD most certainly need CUDA. Without it, AMDs solution is no solution at all for many customers.
AMD could have tanked the price to 1/10th, and nvidia would still hold a majority of the marketshare.

You do realize why AMDs workstation GPU is doing badly? Even through they keep tanking the price? AMD most certainly need CUDA. Without it, AMDs solution is no solution at all for many customers.
AMD could have tanked the price to 1/10th, and nvidia would still hold a majority of the marketshare.

iirc AMD should have some old patents for SMT (pre 2000 I believe). I really doubt they would license Hyper-threading.
HBM/HMC will serve for fast bandwidth (like a L4 cache), and DDR4 will serve for capacity.
KNL will have both HMC and DDR4.

a tumblr social warrior is the term we are looking for. What a time to be alive, really -.-