Nothing too surprising here. Games that previously scaled well in Crossfire
saw excellent performance with the X2, even faster than 8800GT SLI in some
cases. Games that previously didn't scale well, saw poor performance. http://www.fpslabs.com/reviews/video...3870-x2-review
Interestingly, all the Crossfire functions are managed by the large bridge
chip onboard. The drivers see a single card. This could be a blessing
(dual-monitors function transparently) or a curse (the odd game that shows
graphics corruption in Crossfire mode).
--
"War is the continuation of politics by other means.
It can therefore be said that politics is war without
bloodshed while war is politics with bloodshed."
"First of One" <root@127.0.0.1> wrote in message
news:46WdnQ3HyMtHCQvanZ2dnUVZ_j6dnZ2d@giganews.com ...
> Nothing too surprising here. Games that previously scaled well in
> Crossfire saw excellent performance with the X2, even faster than 8800GT
> SLI in some cases. Games that previously didn't scale well, saw poor
> performance.
> http://www.fpslabs.com/reviews/video...3870-x2-review
>
> Interestingly, all the Crossfire functions are managed by the large bridge
> chip onboard. The drivers see a single card. This could be a blessing
> (dual-monitors function transparently) or a curse (the odd game that shows
> graphics corruption in Crossfire mode).
>
> --
> "War is the continuation of politics by other means.
> It can therefore be said that politics is war without
> bloodshed while war is politics with bloodshed."
>
Those COJ scores are pretty impressive. It would have been nice if ATI/AMD
had the R680 lineup available at the time of the 8800 series launch. Better
late than never I guess.
Having the R680 (HD3870 X2) at the time of the 8800 series launch would have
been an absolutely impossible leap. Remember, despite being late, ATi/AMD
still rushed the HD2900XT out the door, with certain parts of the silicon
not working correctly (hardware AA resolve, UVD...).
To the benefit of the consumer though, the 3870 was mainstream-priced and
available in reasonable quantities. Remember the X850XT Platinum-rare
Edition cards going for $500? Did those things really contribute to healthy
competition?
--
"War is the continuation of politics by other means.
It can therefore be said that politics is war without
bloodshed while war is politics with bloodshed."
"Tony DiMarzio" <tony.dimarzio@gmail.com> wrote in message
news:zYmdnWbRqPPQTAvanZ2dnUVZ_v6rnZ2d@comcast.com. ..
> Those COJ scores are pretty impressive. It would have been nice if ATI/AMD
> had the R680 lineup available at the time of the 8800 series launch.
> Better late than never I guess.
>
> Tony
Oops. I meant the _RV670_ lineup should have been the product lineup
available from the ATI side at 8800 series launch. It would have been a much
more competitive offering from the red team. Then, the R680 should have
followed as the high end part.
Yes, it's a completely unrealistic timeline, given delay-plagued history and
bugged silicon of the HD2900XT (R600). I'm just saying ... it would have
been nice
Either way, I'm not upgrading next (to a card from either player) until
Crysis is playable at 1600x1200 with full/max detail as well as full AA/AF
with a single card solution. When that kind of power is available from a
single card, it will be decision time for me.
Tony
--
Tony DiMarzio tony.dimarzio@gmail.com
"First of One" <root@127.0.0.1> wrote in message
news:qKKdnXgOGaGWRQvanZ2dnUVZ_uOmnZ2d@giganews.com ...
> Having the R680 (HD3870 X2) at the time of the 8800 series launch would
> have been an absolutely impossible leap. Remember, despite being late,
> ATi/AMD still rushed the HD2900XT out the door, with certain parts of the
> silicon not working correctly (hardware AA resolve, UVD...).
>
> To the benefit of the consumer though, the 3870 was mainstream-priced and
> available in reasonable quantities. Remember the X850XT Platinum-rare
> Edition cards going for $500? Did those things really contribute to
> healthy competition?
>
> --
> "War is the continuation of politics by other means.
> It can therefore be said that politics is war without
> bloodshed while war is politics with bloodshed."
>
> "Tony DiMarzio" <tony.dimarzio@gmail.com> wrote in message
> news:zYmdnWbRqPPQTAvanZ2dnUVZ_v6rnZ2d@comcast.com. ..
>> Those COJ scores are pretty impressive. It would have been nice if
>> ATI/AMD had the R680 lineup available at the time of the 8800 series
>> launch. Better late than never I guess.
>>
>> Tony
>
>
On Wed, 23 Jan 2008 11:02:07 -0500, "Tony DiMarzio"
<tony.dimarzio@gmail.com> wrote:
>Oops. I meant the _RV670_ lineup should have been the product lineup
>available from the ATI side at 8800 series launch. It would have been a much
>more competitive offering from the red team. Then, the R680 should have
>followed as the high end part.
>
>Yes, it's a completely unrealistic timeline, given delay-plagued history and
>bugged silicon of the HD2900XT (R600). I'm just saying ... it would have
>been nice
>
>Either way, I'm not upgrading next (to a card from either player) until
>Crysis is playable at 1600x1200 with full/max detail as well as full AA/AF
>with a single card solution. When that kind of power is available from a
>single card, it will be decision time for me.
>
Not too long now for the single-card/single-CHIP solution to this woe.
Won't be inexpensive, but hey buy 2 cards using the same chip and use
the second oine for physics acceleration when not using it to impress
your friends with the fastest desktop graphics (in SLI) on the planet.
John Lewis
>Tony
>
>--
>Tony DiMarzio
>tony.dimarzio@gmail.com
>"First of One" <root@127.0.0.1> wrote in message
>news:qKKdnXgOGaGWRQvanZ2dnUVZ_uOmnZ2d@giganews.co m...
>> Having the R680 (HD3870 X2) at the time of the 8800 series launch would
>> have been an absolutely impossible leap. Remember, despite being late,
>> ATi/AMD still rushed the HD2900XT out the door, with certain parts of the
>> silicon not working correctly (hardware AA resolve, UVD...).
>>
>> To the benefit of the consumer though, the 3870 was mainstream-priced and
>> available in reasonable quantities. Remember the X850XT Platinum-rare
>> Edition cards going for $500? Did those things really contribute to
>> healthy competition?
>>
>> --
>> "War is the continuation of politics by other means.
>> It can therefore be said that politics is war without
>> bloodshed while war is politics with bloodshed."
>>
>> "Tony DiMarzio" <tony.dimarzio@gmail.com> wrote in message
>> news:zYmdnWbRqPPQTAvanZ2dnUVZ_v6rnZ2d@comcast.com. ..
>>> Those COJ scores are pretty impressive. It would have been nice if
>>> ATI/AMD had the R680 lineup available at the time of the 8800 series
>>> launch. Better late than never I guess.
>>>
>>> Tony
>>
>>
>
>
"John Lewis" <john.dsl@verizon.net> wrote in message
news:4798ce44.890444@news.verizon.net...
> On Wed, 23 Jan 2008 11:02:07 -0500, "Tony DiMarzio"
> <tony.dimarzio@gmail.com> wrote:
>
>>Oops. I meant the _RV670_ lineup should have been the product lineup
>>available from the ATI side at 8800 series launch. It would have been a
>>much
>>more competitive offering from the red team. Then, the R680 should have
>>followed as the high end part.
>>
>>Yes, it's a completely unrealistic timeline, given delay-plagued history
>>and
>>bugged silicon of the HD2900XT (R600). I'm just saying ... it would have
>>been nice
>>
>>Either way, I'm not upgrading next (to a card from either player) until
>>Crysis is playable at 1600x1200 with full/max detail as well as full AA/AF
>>with a single card solution. When that kind of power is available from a
>>single card, it will be decision time for me.
>>
>
> Not too long now for the single-card/single-CHIP solution to this woe.
> Won't be inexpensive, but hey buy 2 cards using the same chip and use
> the second oine for physics acceleration when not using it to impress
> your friends with the fastest desktop graphics (in SLI) on the planet.
>
> John Lewis
Are you referring to either of the upcoming 9800GTX and 9800GX2? If so, I
don't think either of those cards will provide the necessary power for the
Crysis holy grail.
Tony
>>--
>>Tony DiMarzio
>>tony.dimarzio@gmail.com
>>"First of One" <root@127.0.0.1> wrote in message
>>news:qKKdnXgOGaGWRQvanZ2dnUVZ_uOmnZ2d@giganews.c om...
>>> Having the R680 (HD3870 X2) at the time of the 8800 series launch would
>>> have been an absolutely impossible leap. Remember, despite being late,
>>> ATi/AMD still rushed the HD2900XT out the door, with certain parts of
>>> the
>>> silicon not working correctly (hardware AA resolve, UVD...).
>>>
>>> To the benefit of the consumer though, the 3870 was mainstream-priced
>>> and
>>> available in reasonable quantities. Remember the X850XT Platinum-rare
>>> Edition cards going for $500? Did those things really contribute to
>>> healthy competition?
>>>
>>> --
>>> "War is the continuation of politics by other means.
>>> It can therefore be said that politics is war without
>>> bloodshed while war is politics with bloodshed."
>>>
>>> "Tony DiMarzio" <tony.dimarzio@gmail.com> wrote in message
>>> news:zYmdnWbRqPPQTAvanZ2dnUVZ_v6rnZ2d@comcast.com. ..
>>>> Those COJ scores are pretty impressive. It would have been nice if
>>>> ATI/AMD had the R680 lineup available at the time of the 8800 series
>>>> launch. Better late than never I guess.
>>>>
>>>> Tony
>>>
>>>
>>
>>
>
Intel acquired Havoc. Physics acceleration on video cards is dead in the
absence of another API.
--
"War is the continuation of politics by other means.
It can therefore be said that politics is war without
bloodshed while war is politics with bloodshed."
"John Lewis" <john.dsl@verizon.net> wrote in message
news:4798ce44.890444@news.verizon.net...
> Not too long now for the single-card/single-CHIP solution to this woe.
> Won't be inexpensive, but hey buy 2 cards using the same chip and use
> the second oine for physics acceleration when not using it to impress
> your friends with the fastest desktop graphics (in SLI) on the planet.
>
> John Lewis
>
"First of One" <root@127.0.0.1> wrote in message
news:kL2dnaevAMbsggTanZ2dnUVZ_vyinZ2d@giganews.com
> Intel acquired Havoc. Physics acceleration on video cards is dead in
> the absence of another API.
It was dead the moment Microsoft decided DirectX Physics would be CPU-based.
Didn't read this too closely four days ago, but isn't CPU-based
DirectX-anything an absolute oxymoron? Should we have DirectX AI while we
are at it? :-)
Or do the industry players foresee CPUs with different feature sets? ("This
game requires DirectX 10.2b-compliant CPU... with Intel Application
Accelerator v3.2.56789 Hotfix or later")
--
"War is the continuation of politics by other means.
It can therefore be said that politics is war without
bloodshed while war is politics with bloodshed."
"DRS" <drs@removethis.ihug.com.au> wrote in message
news:47993c1f$0$17198$5a62ac22@per-qv1-newsreader-01.iinet.net.au...
> "First of One" <root@127.0.0.1> wrote in message
> news:kL2dnaevAMbsggTanZ2dnUVZ_vyinZ2d@giganews.com
>> Intel acquired Havoc. Physics acceleration on video cards is dead in
>> the absence of another API.
>
> It was dead the moment Microsoft decided DirectX Physics would be
> CPU-based.
"First of One" <root@127.0.0.1> wrote in message
news:RdKdneRTGfz0GAPanZ2dnUVZ_uidnZ2d@giganews.com
> "DRS" <drs@removethis.ihug.com.au> wrote in message
> news:47993c1f$0$17198$5a62ac22@per-qv1-newsreader-01.iinet.net.au...
>> "First of One" <root@127.0.0.1> wrote in message
>> news:kL2dnaevAMbsggTanZ2dnUVZ_vyinZ2d@giganews.com
>>> Intel acquired Havoc. Physics acceleration on video cards is dead in
>>> the absence of another API.
>>
>> It was dead the moment Microsoft decided DirectX Physics would be
>> CPU-based.
> Didn't read this too closely four days ago, but isn't CPU-based
> DirectX-anything an absolute oxymoron?
My bad (brain ****). The Direct Physics engine is/will be GPU-based (at
least in the short to medium term; longer term CPU/GPU mergers remain
speculative). My point is that once Microsoft, with its extensive influence
on gaming via DirectX, decided its physics engine would not be PPU-based,
the likes of Ageia were dead in the water.
> Should we have DirectX AI
> while we are at it? :-)