<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: CG Channel&#8217;s hardware reviews policy explained</title>
	<atom:link href="http://www.cgchannel.com/2011/10/cg-channels-hardware-reviews-policy-explained/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.cgchannel.com/2011/10/cg-channels-hardware-reviews-policy-explained/</link>
	<description>Community for Entertainment Artists</description>
	<lastBuildDate>Thu, 14 Mar 2013 03:06:54 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.5.1</generator>
	<item>
		<title>By: Razor</title>
		<link>http://www.cgchannel.com/2011/10/cg-channels-hardware-reviews-policy-explained/#comment-24174</link>
		<dc:creator>Razor</dc:creator>
		<pubDate>Sat, 31 Mar 2012 14:53:29 +0000</pubDate>
		<guid isPermaLink="false">http://www.cgchannel.com/?p=28754#comment-24174</guid>
		<description><![CDATA[Adobe&#039;s new Photoshop CS6 now supports OpenCL. Preferences-&gt; Performance-&gt; Advanced Settings-&gt; Use OpenCL.]]></description>
		<content:encoded><![CDATA[<p>Adobe&#8217;s new Photoshop CS6 now supports OpenCL. Preferences-&gt; Performance-&gt; Advanced Settings-&gt; Use OpenCL.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Steve</title>
		<link>http://www.cgchannel.com/2011/10/cg-channels-hardware-reviews-policy-explained/#comment-16102</link>
		<dc:creator>Steve</dc:creator>
		<pubDate>Sun, 13 Nov 2011 11:34:35 +0000</pubDate>
		<guid isPermaLink="false">http://www.cgchannel.com/?p=28754#comment-16102</guid>
		<description><![CDATA[I&#039;d like to see the same model used in all dcc&#039;s. Then it really does become a fairer comparison, as the textures play a large part in these numbers.  

Also I want the model to be freely downloadable, that way people can fairly compare their own rigs, pro or consumer cards. And this will allow people to post their results up.

If one wants to be unbiased and let the community join in, this is the only way to go.]]></description>
		<content:encoded><![CDATA[<p>I&#8217;d like to see the same model used in all dcc&#8217;s. Then it really does become a fairer comparison, as the textures play a large part in these numbers.  </p>
<p>Also I want the model to be freely downloadable, that way people can fairly compare their own rigs, pro or consumer cards. And this will allow people to post their results up.</p>
<p>If one wants to be unbiased and let the community join in, this is the only way to go.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Mark Kauffman</title>
		<link>http://www.cgchannel.com/2011/10/cg-channels-hardware-reviews-policy-explained/#comment-14689</link>
		<dc:creator>Mark Kauffman</dc:creator>
		<pubDate>Tue, 01 Nov 2011 15:52:43 +0000</pubDate>
		<guid isPermaLink="false">http://www.cgchannel.com/?p=28754#comment-14689</guid>
		<description><![CDATA[Thanks for a better explanation, Jason. I posted this review to many of my colleagues and they made several comments about various aspects of the testing, most of which you outlined above. I am glad you plan on expanding your testing procedures further because I feel that Cinebench and SPEC lack any really helpful real world benchmarks. Well, actually, SPEC recently released a basic and pro version of the 3ds Max benchmarks which are much more in line with real world usage of that particular application, specifically 3ds Max 2011. I personally have tried to run it with 2012, but there are some Maxscripting hurtles that I must overcome first to get it to run correctly. This benchmark already needs an update due to the switch to Nitrous and benching based on both shaded and realistic.
Maya is in serious need of an update on this front as well. At my last check, the Maya benchmark was based on 2009. Viewport 2.0 was a huge leap forward like Nitrous and is in need of the appropriate benches to reflect the capabilities here. 
By the way, thanks for pointing out the performance variances between the applications in the multi-application viewport benchmark. You are correct that comparing Nitrous and Viewport 2.0 to the standard OpenGL shading performance in Cinema, Softimage, and others is not a fair or valid comparison. 
As to the call for a comparison to consumer GPUs, I am interested to see what happens. We did some benchmarking of some pro and consumer cards about 6 months ago and ended up with some interesting results. This test was a result of a project where we built a driving simulator and had several types of cards on hand to test in the same system. The simulator was built using the Unreal Dev Kit (UDK). It was interesting that a GeForce 550 GTX Ti would perform much better than a Quadro 5000 in the Unreal Engine, but in 3ds Max, the GeForce performed at ¼ the performance of the Quadro. It makes sense, understanding a little about how the 3d game engines work compared to a 3d application such as 3ds Max or Maya. But, that is a long story…
In any case, I look forward to tests in the future, especially if you come up with a series of new tests that address many of the issues you listed above. Thanks again!]]></description>
		<content:encoded><![CDATA[<p>Thanks for a better explanation, Jason. I posted this review to many of my colleagues and they made several comments about various aspects of the testing, most of which you outlined above. I am glad you plan on expanding your testing procedures further because I feel that Cinebench and SPEC lack any really helpful real world benchmarks. Well, actually, SPEC recently released a basic and pro version of the 3ds Max benchmarks which are much more in line with real world usage of that particular application, specifically 3ds Max 2011. I personally have tried to run it with 2012, but there are some Maxscripting hurtles that I must overcome first to get it to run correctly. This benchmark already needs an update due to the switch to Nitrous and benching based on both shaded and realistic.<br />
Maya is in serious need of an update on this front as well. At my last check, the Maya benchmark was based on 2009. Viewport 2.0 was a huge leap forward like Nitrous and is in need of the appropriate benches to reflect the capabilities here.<br />
By the way, thanks for pointing out the performance variances between the applications in the multi-application viewport benchmark. You are correct that comparing Nitrous and Viewport 2.0 to the standard OpenGL shading performance in Cinema, Softimage, and others is not a fair or valid comparison.<br />
As to the call for a comparison to consumer GPUs, I am interested to see what happens. We did some benchmarking of some pro and consumer cards about 6 months ago and ended up with some interesting results. This test was a result of a project where we built a driving simulator and had several types of cards on hand to test in the same system. The simulator was built using the Unreal Dev Kit (UDK). It was interesting that a GeForce 550 GTX Ti would perform much better than a Quadro 5000 in the Unreal Engine, but in 3ds Max, the GeForce performed at ¼ the performance of the Quadro. It makes sense, understanding a little about how the 3d game engines work compared to a 3d application such as 3ds Max or Maya. But, that is a long story…<br />
In any case, I look forward to tests in the future, especially if you come up with a series of new tests that address many of the issues you listed above. Thanks again!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: ImperfectLink</title>
		<link>http://www.cgchannel.com/2011/10/cg-channels-hardware-reviews-policy-explained/#comment-14645</link>
		<dc:creator>ImperfectLink</dc:creator>
		<pubDate>Mon, 31 Oct 2011 20:57:00 +0000</pubDate>
		<guid isPermaLink="false">http://www.cgchannel.com/?p=28754#comment-14645</guid>
		<description><![CDATA[Thanks for the clarification. I think you&#039;ll see more and more interest in consumer cards for &#039;prosumer&#039; use. The economy being a major reason. It&#039;s a freelance market now and not many are willing to raise their overhead.

I personally have used Quadros, Radeons and Geforce for professional work and have had a couple issues with Quadros being unstable (we ended up with a Radeon in the machine, not by my choice, but because Geforces at the time we too power hungry) and having swapped out three separate Quadros 3k and 4k  FX series, gave up on them.

At home I use a GTX280 that still keeps up in many situations with the Pro cards and have never had issues with artifacting in 3d apps. One has to be a little more careful with newer drivers but that&#039;s about it. In the past, there were many issues with visual fidelity and in some cases, stability using a consumer card but I haven&#039;t seen it in a fair few years (not counting Radeons).

The next issue is that a larger portion of the industry is now creating content for video games which is where the consumer card is a better choice. I guess it really depends where your final scene is being put together, in the application or in the engine.]]></description>
		<content:encoded><![CDATA[<p>Thanks for the clarification. I think you&#8217;ll see more and more interest in consumer cards for &#8216;prosumer&#8217; use. The economy being a major reason. It&#8217;s a freelance market now and not many are willing to raise their overhead.</p>
<p>I personally have used Quadros, Radeons and Geforce for professional work and have had a couple issues with Quadros being unstable (we ended up with a Radeon in the machine, not by my choice, but because Geforces at the time we too power hungry) and having swapped out three separate Quadros 3k and 4k  FX series, gave up on them.</p>
<p>At home I use a GTX280 that still keeps up in many situations with the Pro cards and have never had issues with artifacting in 3d apps. One has to be a little more careful with newer drivers but that&#8217;s about it. In the past, there were many issues with visual fidelity and in some cases, stability using a consumer card but I haven&#8217;t seen it in a fair few years (not counting Radeons).</p>
<p>The next issue is that a larger portion of the industry is now creating content for video games which is where the consumer card is a better choice. I guess it really depends where your final scene is being put together, in the application or in the engine.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
