No problem, but it seems Nvidia PR have dont their jobs and send the information to every sites that they have in their listing, when AMD seems have not .. or the sites dont read mail from AMD ( or just dont care who know ... ) .
.
AMD: Speculation, Rumors, and Discussion (Archive)
- Thread starter iMacmatician
- Start date
- Status
No problem, but it seems Nvidia PR have dont their jobs and send the information to every sites that they have in their listing, when AMD seems have not .. or the sites dont read mail from AMD ( or just dont care who know ... ) .
.
Heh, it's more like AMD PR has done a good job to get even mentioned in Alienware PR piece at all. As I don't see why any one who are willing to spend that kind of money with Alienware "first VR ready" laptop would go with AMD GPU.
Less performance due to to power and heat constrains? Check.
Less performance due to higher CPU utilization against mobile class CPU (read: less potent than desktop), which hamper the system's capability to deliver high FPS required by VR? Check.
Lack of some of VR performance boosting feature such ass multi-projection? Check. (Also check HardOCP tests on some VR games.)
Might as well just get the much cheaper Lenovos with AMD GPU in it.
Less performance due to to power and heat constrains? Check.
Less performance due to higher CPU utilization against mobile class CPU (read: less potent than desktop), which hamper the system's capability to deliver high FPS required by VR? Check.
Lack of some of VR performance boosting feature such ass multi-projection? Check. (Also check HardOCP tests on some VR games.)
Might as well just get the much cheaper Lenovos with AMD GPU in it.
you could be reallly surprised ... Im realy sorry, but i will really not cite hardOCP benchmarks, analysis after his campaing anti-AMD for look credible...
Last edited:
ImSpartacus
Regular
I don't know if that materially changes the situation. Unless AMD is heavily discounting their stuff, Nvidia will be tough to beat in the mobile world.
When we will have the benchmarks of both in hands with this laptops, we will discuss about it . ( Not that it will be better or worst, personally i dont care ... If suddenly the battery life of the Alien 15 is dying in 1 hour , i will let you know if this is due to the AMD gpu ) ..
Haven`t been able to operate the hacked driver here in order to use the BIOS with the tighter timings, so my RX 480 still chugs along at 27.3 MH/sec, BUT! I have been able to figure out this power thingie a bit more with the help of the awesome WattTool from overclock.net. Seems like GPU voltage was tied somehow to memory voltage and as soon as you wanted to go lower than on memory, it just reverted to the higher value.
Am at real 0.88v for the GPU now (1065 MHz) and 0,92v for memory (I suppose controllers, not chips) thanks to a voltage offset. Fan of the reference design spinning at a comfortable 1.700 rpm, keeping the GPU at 69°C. Memory running at 2.225 MHz (x2 if you prefer). Resulting in ~71 watts GPU power use as per GPU-z. That's really cool!
edit: For completeness' sake: http://www.overclock.net/t/1609782/...ith-vrm-monitoring-tweaking-for-rx-400-series
Last edited:
gamervivek
Regular
Zauba has seen a flurry of activity lately with confirmation of C95xxx as Ellsmere(GL?) and a D series PCB that might be for Polaris 11.
There's also a new D series card which is priced very high, probably the counterpart of C993xxx which is ~3x the price of Polaris 10 cards. Though it's even higher than that.
https://www.zauba.com/import-printed+circuit+board+assembly/hs-code-84733030/ip-INHYD4-hs-code.html
There's also a new D series card which is priced very high, probably the counterpart of C993xxx which is ~3x the price of Polaris 10 cards. Though it's even higher than that.
https://www.zauba.com/import-printed+circuit+board+assembly/hs-code-84733030/ip-INHYD4-hs-code.html
Anarchist4000
Veteran
I think it's more memory latency than just bandwidth. A 290 and 480 both hash about the same rate and a Fury X is ~10% higher. That fury would have far more raw bandwidth. Although with the added memory channels of HBM I'd have expected the fury to do a bit better on latency.
D
Deleted member 2197
Guest
New RX 400 Series Cards Outed By AMD?
http://www.hardocp.com/news/2016/08/31/new_rx_400_series_cards_outed_by_amd63/
Didn't we discuss this particular information a month ago?
Edit: Okay, it has changed somewhat:
https://www.reddit.com/r/Amd/comments/4qe7g2/rx490_listed_on_amd_website_q4_release_date/
Edit: Okay, it has changed somewhat:
https://www.reddit.com/r/Amd/comments/4qe7g2/rx490_listed_on_amd_website_q4_release_date/
D
Deleted member 13524
Guest
Yup, same old typo news, just different typo this time.
R9 490 would never exist, since AMD has stated before that the "X" in RX is a performance indicator. It wouldn't make sense to have a RX 460 and R9 490.
IIRC, RX will be used in graphics cards whereas the R7, R5, etc. will be for APUs.
First time I've heard about this. So, an R9 Fury X has „an audio engine“, while R9 Fury and R9 Nano do not?
Albeit a very lax one. AMD said, that the X would henceforth denote their „gaming“ products, defined by ">1.5 TFLOPS and 100 GB/s bandwidth" or "<60 fps@1080p in popular games (Dota 2, LoL etc.).
With this definition, there would be no stopping from having a Hawaii-based SKU branded as a R9 490 - especially if found on a microsite for OEM products. OEM names have existed outside the rules for... ever?
Ext3h
Regular
Ah, never mind. There was this strange anomaly back in the 200 series, that the 260 with X suffix had the TrueAudio DSP enabled, while the ones without didn't (despite being based on the same silicon). Double checked, and this pattern didn't apply to the other chips.
- Status
