Maybe ring-bus is back
(I believe RV570 supported combination of 12 ROPs with 256bit interface)
AMD: Southern Islands (7*** series) Speculation/ Rumour Thread
- Thread starter UniversalTruth
- Start date
(I believe RV570 supported combination of 12 ROPs with 256bit interface)LOL.There are at least 3 or 4 double-confirmed configurations:
No EDRAM this time?
- 384bit GDDR5
- 256bit XDR2
- 384bit QDR DDR5
- 256bit 7Gbps GDDR5
- …
btw 7gbps gddr5 has to be some kind of running joke, it is always "shipping early next year", since 2008
. Maybe it'll be true at some point...Ninjaprime
Regular
LOL.
btw 7gbps gddr5 has to be some kind of running joke, it is always "shipping early next year", since 2008
I would guess its been around, just no one yet wants to make a 7 gbps capable memory controller. IIRC, AMD saying their transition from 4870 --> 5870 memory controller needed twice the logic/wiring, and similar reasoning for fermi's memory controller not being up to snuff with AMDs.
Yeah, that and nobody wants the power consumption of 7Gbps chips either. That would be my guess, anyway.
Of course, eDRAM is mostly used to replace SRAM not DRAM.There are at least 3 or 4 double-confirmed configurations:
No EDRAM this time?
- 384bit GDDR5
- 256bit XDR2
- 384bit QDR DDR5
- 256bit 7Gbps GDDR5
- …
It sounds so stupid if there are rumours about eDRAM
Really? I haven't heard any place where they use eDRAM instead of DRAM
I just know some chips prefer eDRAM against SRAM.
They should be de-coupled (if it's even remotely true which I severely doubt).
It's been used in consoles to reduce the need for a high-bandwidth DRAM bus, not DRAM itself.
A modest bus would still be needed to connect to the high capacity DRAM, at least until we learn how to stack scads of them on a 3d IC.
A modest bus would still be needed to connect to the high capacity DRAM, at least until we learn how to stack scads of them on a 3d IC.
Well last time I checked the only offers even at 6gbps were factory overvolted (1.6V instead of 1.5V). If you read the announcements of those memory companies, you'd get the impression they had 7gbps chips running at 1.35V in 2009 which would certainly help with the power consumption. But of course those announcements are pretty shady - what they wanted to say is "we've got a chip which could maybe run at 7gbps at an undisclosed voltage (and we don't tell you how high it really needs to be because it isn't practically useful anyway) and the same chip can also run at 1.35V (but at half the data rate)"...
So if they could make 1.5V 7gbps chips they wouldn't require more power than what amd is currently using (at 1.6V).
I don't think the difference in power consumption will be much more than a linear scale up. If so, it'd be a simple choice between power and performance. One that many would be willing to make.Alexko said:
A bigger argument against the 7Mbps is that the performance increase doesn't scale as expected. There is a threshold (5.5Mbps?) above which addition restrictions are imposed wrt bank access patterns, which can result in performance hit. It was mentioned in an Anandtech article, but I can't find it right now. The result would be that, e.g. 6Mbps will be a net negative and you have to go higher to recoup that loss.
I thought PCIe 3 was end of the line.rpg.314 said:
They find ways to improve, don't worry.
Ninjaprime
Regular
I seem to recall some hynix pdf slides that gave two possible branches for the future, one refered to GDDR5+ scaling up 32gbps, which would be 8ghz. So I think the possiblity is there, but with no demand for even their 7ghz products yet, they have no real reasons to launch anything. Or perhaps they are taking the other branch which refered to some new memory system entirely.
I just see no evidence the memory manufacturers can actually produce 7gbps chips, despite claiming to be able to 3 years ago.
You'd certainly think they'd have 6gbps chips which were using standard 1.5V voltage instead of the factory overvolted 1.6V parts if it's so easy. I bet there IS demand for such such chips (as well as higher speed grade 1.35V parts for mobile products).
Noone is interested in parts needing even higher voltages (and hence with power consumption going through the roof) but I think claming no demand for 7gbps parts is silly (well if it actually can be made to work at standard voltage that is).
I know this info is to be taken with a truck load of salt, but the wiki page for Southern Islands has changed.
7970: 2048:128:48 where 48 = ROPs. This definitely said 64 last week. Is someone in the know already?
I don't know why, but my expectation for a 7970 just cratered...
7970: 2048:128:48 where 48 = ROPs. This definitely said 64 last week. Is someone in the know already?
I don't know why, but my expectation for a 7970 just cratered...
D
Deleted member 13524
Guest
So any news about this rumoured event that should happen today in London?
It won't happen? Will it happen later?
I'm eagerly awaiting news about this launch, not only because I'm curious but also because I have a graphics card purchase pending.
BTW, I saw this:
Also, two new laptops have in the price lists appeared with SI GPUs:
- ASUS X53TK-SX058V w/ A6 3420M + Radeon HD 7670M for ~620€ (wow @ price point)
- HP Envy 17-3030ew w/ i7-2670QM + HD7690M for ~1200€
However, I haven't seen any specific rumour about what will be the HD76xxM GPUs, so there's a big chance they're just renamed HD67xxM Whistlers (VLIW5, 480sp, 24TMUs, 8 ROPs).
It won't happen? Will it happen later?
I'm eagerly awaiting news about this launch, not only because I'm curious but also because I have a graphics card purchase pending.
BTW, I saw this:
Also, two new laptops have in the price lists appeared with SI GPUs:
- ASUS X53TK-SX058V w/ A6 3420M + Radeon HD 7670M for ~620€ (wow @ price point)
- HP Envy 17-3030ew w/ i7-2670QM + HD7690M for ~1200€
However, I haven't seen any specific rumour about what will be the HD76xxM GPUs, so there's a big chance they're just renamed HD67xxM Whistlers (VLIW5, 480sp, 24TMUs, 8 ROPs).
Similar threads
