Almost Official - NV28/18 only adds AGP 8x support
- Thread starter Pav_37
- Start date
Simpel question: Why in the world would nVidia miss out on customers by simply naming it Ti4600 + "something".
That indicates that it's basically the same product as the normal Ti 4600 in everyones ears right? (Sure does in mine anyways.)
It's sorta like there's a Radeon 9000 and a Radeon 9000 PCI me thinks. If the normal Radeon 9000 (we're leaving pro out of this) was that much better wouldn't they just call it well, Radeon 9200 or Radeon 9000 Pro or something? Especially if they launched it later.
Otherwise The Ti4200 "AGP8X edition" seems like a VERY strange product. If your assumptions (that's what I'd label them anyway) are true Ti 4200 should be very much on par with a Ti4400 so why still call it a Ti4200 if you catch my drift.
Why name a faster product with the same name as a slower product? Basically that's my point and I think it speaks clearer than any jibberish or tech mumbo jumbo.
The Ti numbers correspond (IMO) to the "GTS, "Pro" and "Ultra".
The "GeForce 2 Ultra" was called "GeForce 2 Ultra" and not "GeForce 2 Pro, Takes-slightly-better-advantage-of-the-AGP-port version" because of a reason.
Namely: sales. And you gotta give drecit where credit is due and nVidias PR department deserves loads of it (credit that is). I don't think they'd let such a foolish naming scheme destroy potential sales.
That indicates that it's basically the same product as the normal Ti 4600 in everyones ears right? (Sure does in mine anyways.)
It's sorta like there's a Radeon 9000 and a Radeon 9000 PCI me thinks. If the normal Radeon 9000 (we're leaving pro out of this) was that much better wouldn't they just call it well, Radeon 9200 or Radeon 9000 Pro or something? Especially if they launched it later.
Otherwise The Ti4200 "AGP8X edition" seems like a VERY strange product. If your assumptions (that's what I'd label them anyway) are true Ti 4200 should be very much on par with a Ti4400 so why still call it a Ti4200 if you catch my drift.
Why name a faster product with the same name as a slower product? Basically that's my point and I think it speaks clearer than any jibberish or tech mumbo jumbo.
The Ti numbers correspond (IMO) to the "GTS, "Pro" and "Ultra".
The "GeForce 2 Ultra" was called "GeForce 2 Ultra" and not "GeForce 2 Pro, Takes-slightly-better-advantage-of-the-AGP-port version" because of a reason.
Namely: sales. And you gotta give drecit where credit is due and nVidias PR department deserves loads of it (credit that is). I don't think they'd let such a foolish naming scheme destroy potential sales.
Joe DeFuria
Legend
Or maybe some people are just fishing for something they hope isn't real?
My speculation: The NV18/28's relationship to the NV17/NV25 will be similar to the GeForce2 "Pro" vs. the GeForce2 GTS. Some minor tweakage to the base design to increase yield / lower cost. (That, plus throw in AGP 8X support).
End result: we'll get virtually the same 4200 and 4600 products we know today, but with AGP 8x support. The biggest benefit to nVidia is probably slightly lower manufacturing cost compared to the NV17/25, which will help extend the life of those products a little bit. So, these chips are probably more significant to nVidia than they are to the end users.
Joe DeFuria
Legend
Yes, uh, erhm....I could've sworn you were speculating on increased clocks, cache sizes, etc., and how NV18/28 must certainly be more effectively than just NV17/NV25 with AGP 8x support....
Joe DeFuria said:
I'm talking about the "these chips are probably more significant to nVidia than they are to the end users"
And besides, who found the "GeForce4 Ti 4600 with AGP8X &
GeForce4 Ti 4200 with AGP8X" anyway? I've installed the drivers myself and looked at the new .inf and the old .inf (they left it there) and none of this is mentioned (only NV18/NV28).
BoddoZerg said:
I think you mean the fanboys on this board...!
kid_crisis
Newcomer
Those are the numbers that were presented at the CC on July 11th, Nvidia's yearly shareholder meeting. The numbers were clearly visible on the slide that was presented there, and were also verbally stated by Jen Hsun as he was presenting the slide.
NV30 - 120M
NV28 - 86M
NV18 - 81M
It's become clear (in retrospect) that the 81M transistors for the NV18 is the "crush" version, which is the Northbridge of the Nforce2.
The night of the CC I documented my notes and posted them here:
http://forums.delphiforums.com/thegraphicxlair/messages?msg=595.1
http://forums.delphiforums.com/thegraphicxlair/messages?msg=595.2
These were subsequently copied over to the Motley Fool, and then migrated to other places on the net.
NV30 - 120M
NV28 - 86M
NV18 - 81M
It's become clear (in retrospect) that the 81M transistors for the NV18 is the "crush" version, which is the Northbridge of the Nforce2.
The night of the CC I documented my notes and posted them here:
http://forums.delphiforums.com/thegraphicxlair/messages?msg=595.1
http://forums.delphiforums.com/thegraphicxlair/messages?msg=595.2
These were subsequently copied over to the Motley Fool, and then migrated to other places on the net.
Is there also an integrated NV28(super Xbox), or would that be impractical due to the limited memory bandwidth for cpu+gpu?
I am very skeptical that they would put 23 million extra transistors onto NV28. In the past, high-end nVidia chips have always come with huge amounts of hype, while low-end chips (like NV17) generally have little or no hype. The total lack of hype for NV28 would imply that it is not a high end part, and it seems crazy to put 23m more transistors into GeForce4 when it's not going to be a high-end product.
I'd agree that NV28 is probably an AGP8x GeForce4, perhaps clocked higher or exactly identical to current GF4s. (Same with NV18 and GF4MX) The only thing that worries me about that hypothesis is that I wouldn't think nVidia would have to come up with a new "NV" name just for AGP8x. For previous nV cards the PCI and AGP versions have always been the same "NV" number, and the difference between PCI and AGP4x is much greater than the difference between AGP4x and AGP8x.
I am very skeptical that they would put 23 million extra transistors onto NV28. In the past, high-end nVidia chips have always come with huge amounts of hype, while low-end chips (like NV17) generally have little or no hype. The total lack of hype for NV28 would imply that it is not a high end part, and it seems crazy to put 23m more transistors into GeForce4 when it's not going to be a high-end product.
I'd agree that NV28 is probably an AGP8x GeForce4, perhaps clocked higher or exactly identical to current GF4s. (Same with NV18 and GF4MX) The only thing that worries me about that hypothesis is that I wouldn't think nVidia would have to come up with a new "NV" name just for AGP8x. For previous nV cards the PCI and AGP versions have always been the same "NV" number, and the difference between PCI and AGP4x is much greater than the difference between AGP4x and AGP8x.
It is indeed very interesting that no information has leaked about NV28 or NV18 other then the fact that it will have AGP 8x support and some number of extra transistors over NV25. At the same time so much hype and marketing BS has been released about NV30. Don't you find it a bit weird that noone is talking about NV18/NV28? No single fan nvidia website, no major hardware site - total silence. Weird huh? NV30 is still so far away yet we already heard a lot about it (nothing concrete though I guess). Makes you wonder............
I guess the latest news about the naming of NV18/NV28 in a way tells us that perhaps there's nothing to talk about? I mean if its only Geforce4 + AGP 8x - who cares? lol - so in a way maybe that explains lack on Marketing PR BS about nv18/nv28 and lack of info leaks in that area.
I guess the latest news about the naming of NV18/NV28 in a way tells us that perhaps there's nothing to talk about?
I mean if its only Geforce4 + AGP 8x - who cares? lol - so in a way maybe that explains lack on Marketing PR BS about nv18/nv28 and lack of info leaks in that area.