vidi ima i ova prica 
http://wccftech.com/article/radeon-rx-480-reducing-voltage-increasing-efficiency/
http://wccftech.com/article/radeon-rx-480-reducing-voltage-increasing-efficiency/
Cudi me da Titana nisu opremili sa kompletnim cipom ako vec u Quadro ide.
Ipak je Quadro bitniji proizvod, a čip sa malo škarta u Titana. Ko zna možda se GP102 pojavi u još nekoj varijanti.
Update 07/25: NVIDIA has given us a few answers to the question above. We have confirmation that the FP64 and FP16 rates are identical to GP104, which is to say very slow, and primarily there for compatibility/debug purposes. With the exception of INT8 support, this is a bigger GP104 throughout.
Meanwhile we have a die size for GP102: 471mm2, which is 139mm2 smaller than GP100. Given that both (presumably) have the same number of FP32 cores, the die space savings and implications are significant. This is as best of an example as we're ever going to get on the die space cost of the HPC features limited to GP100: NVLInk, fast FP64/FP16 support, larger register files, etc. By splitting HPC and graphics/inference into two GPUs, NVIDIA can produce GP102 at what should be a significantly lower price (and higher yield), something they couldn't do until the market for compute products based on GP100 was self-sustaining.
Finally, NVIDIA has clarified the branding a bit. Despite GeForce.com labeling it "the world’s ultimate graphics card," NVIDIA this morning has stated that the primary market is FP32 and INT8 compute, not gaming. Though gaming is certainly possible - and I fully expect they'll be happy to sell you $1200 gaming cards - the tables have essentially been flipped from the past Titan cards, where they were treated as gaming first and compute second. This of course opens the door to a proper GeForce branded GP102 card later on, possibly with neutered INT8 support to enforce the market segmentation.
Update 07/25: NVIDIA has given us a few answers to the question above. We have confirmation that the FP64 and FP16 rates are identical to GP104, which is to say very slow, and primarily there for compatibility/debug purposes. With the exception of INT8 support, this is a bigger GP104 throughout.
Meanwhile we have a die size for GP102: 471mm2, which is 139mm2 smaller than GP100. Given that both (presumably) have the same number of FP32 cores, the die space savings and implications are significant. This is as best of an example as we're ever going to get on the die space cost of the HPC features limited to GP100: NVLInk, fast FP64/FP16 support, larger register files, etc. By splitting HPC and graphics/inference into two GPUs, NVIDIA can produce GP102 at what should be a significantly lower price (and higher yield), something they couldn't do until the market for compute products based on GP100 was self-sustaining.
Finally, NVIDIA has clarified the branding a bit. Despite GeForce.com labeling it "the world’s ultimate graphics card," NVIDIA this morning has stated that the primary market is FP32 and INT8 compute, not gaming. Though gaming is certainly possible - and I fully expect they'll be happy to sell you $1200 gaming cards - the tables have essentially been flipped from the past Titan cards, where they were treated as gaming first and compute second. This of course opens the door to a proper GeForce branded GP102 card later on, possibly with neutered INT8 support to enforce the market segmentation.
- ovo je prvi Titan koji nema dupliranu memoriju (12GB ce biti standardna memorija za GTX 1080Ti model, prosli Titan X je imao 12GB dok je GTX 980Ti imao 6GB)
Dakle sad ide prica da je ovaj Titan X primarno compute karta pa tek onda gaming.
GP102, koji je bukvalno GP104 +50% (isto kao sto je GM200 bio u odnosu na GM204) definitivno ide u sledeci high-end Geforce (GTX 1080Ti)
Ovako stoji za Quadro P6000 na sajtu nVidia-e:
QUADRO P6000 SPECS
CUDA Parallel-Processing Cores 3840
GPU Memory 24 GB GDDR5X
FP 32 Performance 12 TFLOPs
Max Power Consumption 250 W
Graphics Bus PCI Express 3.0 x 16
Display Connectors DP 1.4 (4) DVI-D (1) Optional Stereo (1)
Form Factor 4.4" H x 10.5" L Dual Slot
http://www.nvidia.com/object/quadro-graphics-with-pascal.html
Nekako mi jadno deluje Titan X koji nece imati ceo GP102...
a sto su klosari stavili na sliku cip sa HBM memorijom![]()
I pretpostavio sam da nece biti zvanicnih review-a za Titana jer ga sada vise ne prodaju kao primarno gaming kartu (izbacili su Geforce iz naziva karte).
Novi titan nema nikakvog smisla i neće se prodavati kao prošli modeli.
Koje bi specifikacije 1080ti imao i koju cenu pa da bude bitan?
Mislis 30% + ?
a dobro ne znam racunati znaci oko 30% pa ocekivano, nista po meni veliko, niti bitno...jos nema preko 60 fps za 4k, znaci sledeci titan ce to omoguciti da bude full uzivanje![]()
I te kako je rezultat impresivan. Karta odvaljuje a jos nije potpun cip.
Pogledaj samo razliku u frejmovima.
Witcher 3, snaznija od GTX 1080 za 16 frejmova, Rise of the Tomb Raider - razlika 11 frejmova,
Far Cry Primal - razlika 14 frejmova.. Kida. A gde je overklok?![]()
ja dok ne vidim 75 average fps na 4k meni je ovo sve slabo, to je neka granica da ne pada puno ispod 60, a karta je dobra u svakom slucaju to stoji ali nije za 4k full uzivanje preko 60 fps...
Follow along with the video below to see how to install our site as a web app on your home screen.
Napomena: this_feature_currently_requires_accessing_site_using_safari