Another Ray Tracing demo from Intel

T2k

Golden Member
Feb 24, 2004
1,665
5
81
Video is nothing new... but some comments are funny:

EgoistXIV (4 hours ago) -1 Reply | Spam
Damn German? pervert. Check his face when he first sees the barbie doll.
(...)
1vinas (5 hours ago) 0 Reply | Spam
If Daniel calls that pushing weights hard, I? don't want to see what raytracing will do. Ever.

:D
 

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
This kind of advertising absolutely does not help Intel's cause. Why would they ever authorize this? Embarrassing.
 

alyarb

Platinum Member
Jan 25, 2009
2,425
0
76
yeah, it's definitely a bad video. plus they are just showing close-ups of how slow raytracing is going to be on larrabee.

also genx87 below is right. the wave action is so uniform so it just looks like ribbed glass. terrible.
 

Ben90

Platinum Member
Jun 14, 2009
2,866
3
0
I like the guys raytracing description.... Was raycasting def not tracing
 

nitromullet

Diamond Member
Jan 7, 2004
9,031
36
91
Originally posted by: Idontcare
This kind of advertising absolutely does not help Intel's cause. Why would they ever authorize this? Embarrassing.

Agreed. It looks like Intel has nothing really impressive to show us about Larrabee.
 

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
Originally posted by: nitromullet
Originally posted by: Idontcare
This kind of advertising absolutely does not help Intel's cause. Why would they ever authorize this? Embarrassing.

Agreed. It looks like Intel has nothing really impressive to show us about Larrabee.

Anyone buying into the prospects of this being a sand-bag attempt?

It's crossed my mind but the video is just so bad (combined with the actual IDF press materials datum point) that I'm actually convinced this is legitimately painting an accurate picture of where Larrabee is at.
 

bryanW1995

Lifer
May 22, 2007
11,144
32
91
maybe this is the why gelsinger is gone. of course, if intel's best and brightest can't make it work, maybe they could snag some former ati engineers?

it would probably be cheaper at this point to initiate a hostile takeover of nvidia. huang probably wouldn't go quietly into the night, hence the "hostile" aspect, but with a market cap of only $7 billion chipzilla could easily afford it. jen hsun would probably even thank them later since it would provide him a perfect exit strategy and probably bump the share price quite a bit.
 

T2k

Golden Member
Feb 24, 2004
1,665
5
81
It's the same crap they showed everywhere - fairly unimpressive TBH.
 

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
Originally posted by: bryanW1995
maybe this is the why gelsinger is gone. of course, if intel's best and brightest can't make it work, maybe they could snag some former ati engineers?

it would probably be cheaper at this point to initiate a hostile takeover of nvidia. huang probably wouldn't go quietly into the night, hence the "hostile" aspect, but with a market cap of only $7 billion chipzilla could easily afford it. jen hsun would probably even thank them later since it would provide him a perfect exit strategy and probably bump the share price quite a bit.

If history is any indicator of how Intel will likely deal with NV and AMD in the graphics market I'd say look at how they penetrated the server/enterprise/big-iron markets and look at what their strategy did to the then existing lead players.

Intel pretty much just left them to their own devises, allowing their management to manage their respective companies right into the ground, and Intel just came along afterwards and picked over the carcasses to buy up just the IP and assets they wanted for pennies on the dollar.

DEC is a good example.
 

Genx87

Lifer
Apr 8, 2002
41,091
513
126
Originally posted by: Idontcare
Originally posted by: bryanW1995
maybe this is the why gelsinger is gone. of course, if intel's best and brightest can't make it work, maybe they could snag some former ati engineers?

it would probably be cheaper at this point to initiate a hostile takeover of nvidia. huang probably wouldn't go quietly into the night, hence the "hostile" aspect, but with a market cap of only $7 billion chipzilla could easily afford it. jen hsun would probably even thank them later since it would provide him a perfect exit strategy and probably bump the share price quite a bit.

If history is any indicator of how Intel will likely deal with NV and AMD in the graphics market I'd say look at how they penetrated the server/enterprise/big-iron markets and look at what their strategy did to the then existing lead players.

Intel pretty much just left them to their own devises, allowing their management to manage their respective companies right into the ground, and Intel just came along afterwards and picked over the carcasses to buy up just the IP and assets they wanted for pennies on the dollar.

DEC is a good example.

While this is true. There were also other forces at work. The high cost of R&D for those RISC processors. Entrance of x86 into that realm. Other market forces.

I think the 3d market is a bit different than the server market. Intel cant sell an integrated chipset to the masses to play games or run HPC applications like they could x86 chips in the low and middle server market. In the server market Intel could sell a 4p low end server that performed well for a fraction of the cost of a RISC server. If the performance wasnt upto snuff cluster 4 of them and it is still a fraction of the cost with increased reliability and as good or better performance. Graphics people cant cluster and they demand results. For instance intel could give me their integrated crap for free and I wouldnt take it to play any game made within the past 5 years.

And lets not forget Intels own x86 strategy has basically destroyed all hope for the Itanium to ever be relevant. And I suspect once Larrabee and Fermi show up Itanium could be irrelevant in the only market it has ever seen any kind of success.

That said, I dont see any regulatory body allowing Intel to consume the market leader in discrete graphics that also holds about 20% of the overall market. That would place Intel at about 80% of the overall market and a commanding high margin product in discrete graphics. Not to mention a complete lock on the HPC market for GPUs.
 

Nemesis 1

Lifer
Dec 30, 2006
11,366
2
0
Originally posted by: Idontcare
Originally posted by: nitromullet
Originally posted by: Idontcare
This kind of advertising absolutely does not help Intel's cause. Why would they ever authorize this? Embarrassing.

Agreed. It looks like Intel has nothing really impressive to show us about Larrabee.

Anyone buying into the prospects of this being a sand-bag attempt?

It's crossed my mind but the video is just so bad (combined with the actual IDF press materials datum point) that I'm actually convinced this is legitimately painting an accurate picture of where Larrabee is at.


Man you have changed . DO ya see any lies here? NV just doing a retap out on A2.

What intel showing is at least the truth. NV is retapping A2 and A2 won't be good . I will wager $100 that NV does an A3 at least befor they ship . ANY TAKERS ANYONE . I will send my money to AT they can pay the winner. If its allowed Put up.

 

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
Originally posted by: Nemesis 1
Originally posted by: Idontcare
Originally posted by: nitromullet
Originally posted by: Idontcare
This kind of advertising absolutely does not help Intel's cause. Why would they ever authorize this? Embarrassing.

Agreed. It looks like Intel has nothing really impressive to show us about Larrabee.

Anyone buying into the prospects of this being a sand-bag attempt?

It's crossed my mind but the video is just so bad (combined with the actual IDF press materials datum point) that I'm actually convinced this is legitimately painting an accurate picture of where Larrabee is at.


Man you have changed . DO ya see any lies here? NV just doing a retap out on A2.

What intel showing is at least the truth. NV is retapping A2 and A2 won't be good . I will wager $100 that NV does an A3 at least befor they ship . ANY TAKERS ANYONE . I will send my money to AT they can pay the winner. If its allowed Put up.

Changed? I sure as hell hope so, I try and reinvent myself all the time. Anything less would be to break character.

At any rate what lies am I supposed to be seeing exactly? (or what false-lies do you think I am imagining seeing?)

We got Cypress in stock and shipping from newegg, we got youtube videos showing how craptastic the larrabee program is at the moment while the last guy to have been seen physically holding larrabee's in his hand has left the company, and we got NV's ceo holding up mocked-up products of something we are supposed to believe exists but they haven't found the energy the desire or the time to bother with releasing a stock photo of the internal and functioning cards they do have on hand, be it one of their 7 samples or one of the hundreds they'd just as soon we believe them to have.

Does that about sum up how Q3/09 ended?
 

v8envy

Platinum Member
Sep 7, 2002
2,720
0
0
Hrm. That demo is certainly impressive, and at least as good as anything I saw in Morrowind.

In 2003.

Consider me as part of the headscratching crowd. Who is the target audience for this? Did Intel have any say re: this video going public? If so, that decision maker needs to be marched out of the office ASAP. Makes me wonder if the reason we haven't seen an equivalent NV Tesla demo is because it would be equally uninspiring. Silence and resulting doubt definitely beats being convinced a product is useless.
 

Nemesis 1

Lifer
Dec 30, 2006
11,366
2
0
I think the Title is BS . Its not really a RT demo . Its talking about the ray casting and shadow mapping with raycast. It was interesting but Old . . IDC is what i meant I thought I was Clear . Is No lies in that demo Zero . I like how intel is sand bagging , But if NV takes a B1 to release Intel may just release anyway.
 

T2k

Golden Member
Feb 24, 2004
1,665
5
81
Originally posted by: Idontcare
Originally posted by: bryanW1995
maybe this is the why gelsinger is gone. of course, if intel's best and brightest can't make it work, maybe they could snag some former ati engineers?

it would probably be cheaper at this point to initiate a hostile takeover of nvidia. huang probably wouldn't go quietly into the night, hence the "hostile" aspect, but with a market cap of only $7 billion chipzilla could easily afford it. jen hsun would probably even thank them later since it would provide him a perfect exit strategy and probably bump the share price quite a bit.

If history is any indicator of how Intel will likely deal with NV and AMD in the graphics market I'd say look at how they penetrated the server/enterprise/big-iron markets and look at what their strategy did to the then existing lead players.

Intel pretty much just left them to their own devises, allowing their management to manage their respective companies right into the ground, and Intel just came along afterwards and picked over the carcasses to buy up just the IP and assets they wanted for pennies on the dollar.

DEC is a good example.

Is this the same "enterprise/big iron" aka high-end market where Intel is literally a no-show, even after sinking zillions into its utterly dead IA64 project?

Nonsense... Intel might have acquired this and that but hardly even been considered as a real option when someone needed scaling and I mean beyond the corporate stuff (like 64, 128-way and higher.) Sun or IBM, yes, Intel, sorry, thanks but no, thanks.
 

BenSkywalker

Diamond Member
Oct 9, 1999
9,140
67
91
It's crossed my mind but the video is just so bad (combined with the actual IDF press materials datum point) that I'm actually convinced this is legitimately painting an accurate picture of where Larrabee is at.

They are showing the potential of real time ray tracing. That potential is extremely low. It isn't Intel's fault on a technology front, it was a shockingly dumb idea from day one.

Is this the same "enterprise/big iron" aka high-end market where Intel is literally a no-show, even after sinking zillions into its utterly dead IA64 project?

Intel destroyed the 'big iron' market altogether. It used to consist of billions a year going to makers of enterprise class solutions. Now, most of those setups run x86 based hardware. There is still a very small market segment above x86, but it is a small fraction of what it was years ago.

BTW- This is not ray casting. Ray casting works as a visibility check for depth sort, it doesn't work to handle reflections or light modelling at all.
 

Nemesis 1

Lifer
Dec 30, 2006
11,366
2
0
Originally posted by: BenSkywalker
It's crossed my mind but the video is just so bad (combined with the actual IDF press materials datum point) that I'm actually convinced this is legitimately painting an accurate picture of where Larrabee is at.

They are showing the potential of real time ray tracing. That potential is extremely low. It isn't Intel's fault on a technology front, it was a shockingly dumb idea from day one.

Is this the same "enterprise/big iron" aka high-end market where Intel is literally a no-show, even after sinking zillions into its utterly dead IA64 project?

Intel destroyed the 'big iron' market altogether. It used to consist of billions a year going to makers of enterprise class solutions. Now, most of those setups run x86 based hardware. There is still a very small market segment above x86, but it is a small fraction of what it was years ago.

BTW- This is not ray casting. Ray casting works as a visibility check for depth sort, it doesn't work to handle reflections or light modelling at all.

You best watch that video again . Sky walker . did ya see how that shadow was created . the cast ray could ony be sent recieved one direction so it new were shadow was . I thought that was raycasting had it gone both to the reciever and back to the sender no shadow .

 

lalorlor

Junior Member
Oct 17, 2009
1
0
0
Originally posted by: BenSkywalker
BTW- This is not ray casting. Ray casting works as a visibility check for depth sort, it doesn't work to handle reflections or light modelling at all.

OMG, you don´t have the slightest idea of what ray casting is.