backtop

Blog: Hardware DailyTech Digest: RyderMark
Kristopher Kubicki (Blog) - February 19, 2007 1:26 PM
Print E-mail del.icio.us 31 comment(s) - last by James Holden.. on Mar 5 at 11:37 PM

The RyderMark "screenshot" that was published in July 2006
My take on RyderMark's newest incarnation

Fudo posted something about Rydermark today, and the hate mail is already starting to kick up.

In a previous article, we pretty thoroughly debunked the authenticity of Fudo's "screenshots."  In the original article, Fudo published an expose claiming NVIDIA would not allow developers to use 24-bit or 32-bit precision -- whatever that means -- inside the benchmark.  Fudo then published the screenshots of an ATI and an NVIDIA card side by side.

Our rebuttal was to disprove his analysis of this situation based on the two images, at least one of which was just a Photoshop manipulation.  I have my doubts about Candella as well, but the gist of that article -- Fudo's accusations that NVIDIA was pulling some sort of scam with developers -- was untrue given the evidence.

So here's the gauntlet: If Fudo or Candella can supply me with a version of the RyderMark benchmark by midnight tonight, I will donate $1,000 to a charity of Fudo's choosing.  The software must show, in full-motion video, all three screenshots displayed in the article Fudo published today.  I'll donate another $1,000 if it can display the images from the RyderMark "screenshots" published in July 2006.

Update 2/20/2007: Neither individual was able to provide me with the software even after direct requests.  Fuad declined to send me the binary and Candella did not respond to my emails.

Update 2/20/2007: Ryan Shrout from PC Perspective has uploaded the teaser video for the benchmark on Fileshack.  I'm beginning to think this whole thing has been a horrible hoax played on Fudo.

Comments     Threshold


This article is over a month old, voting and posting comments is disabled
No one noticed this yet?
By Chillin1248 (blog) on 2/20/2007 2:27:07 AM , Rating: 3
Surprising that people are paying so much attention to the pictures and not to the Spec printout, look at the facts:

1)- He claims and the spec sheet claims that he has a QuadFX system with four cores, but why the hell isn't Multi-Threading enabled?

2)- For a graphics benchmark, it sure talks a lot about the CPU data and NOTHING and the GPU data. It doesn't even say which GPU is in there...

3)- Why does the OS description say:

Microsoft Windows XP Sevice pacl

I mean is it really such a BETA that it can't even say Service Pack 2?

And finally my favorite:

Why would S3 be sponsering a graphic benchmark that the fastest and most expensive cards today can't manage, never mind their own low-end cards?

-------
Chillin




RE: No one noticed this yet?
By Le Québécois on 2/20/2007 3:02:20 AM , Rating: 2
I saw that but to be honest I was more amaze by the benchmark results. I know he claims that the benchmark look better in action but still how could something that ugly can perform so badly on a Quad FX with a GeForce 8800?

Is this a new trend to come with benchmarks in the future? Poorly writen code so we can see how our computers will perfom on a bad engine?

Kris, any follow up on that "give me a video and I'll give you $1000"?


RE: No one noticed this yet?
By KristopherKubicki (blog) on 2/20/2007 3:29:30 AM , Rating: 2
Not a peep from Candella, though they have told my colleages they will release a video within 10 days.

Fudo claims he will post a video shortly as well, but no one has been able to provide me of a binary.


RE: No one noticed this yet?
By Chillin1248 (blog) on 2/20/2007 1:42:28 PM , Rating: 3
My god, the INQ doesn't stop shooting themselves in the foot. Read this article posted today which claims that Nvidia has no WHQL Vista driver and that AMD/ATI will beat them:

http://uk.theinquirer.net/default.aspx?article=377...

Now please someone tell him that Nvida DOES HAVE a Vista WHQL driver (100.65):

32-bit:
http://www.nvidia.com/object/winvista_x86_100.65.h...

64-bit:
http://www.nvidia.com/object/winvista_x64_100.65.h...

And also unlike he claimed in the article, this driver does support the 8800 series in SLI.
-------
Chillin


RE: No one noticed this yet?
By KristopherKubicki (blog) on 2/20/2007 1:55:43 PM , Rating: 2
Their driver release dates and embargo dates are wrong anyway.


By KristopherKubicki (blog) on 2/20/2007 3:13:25 PM , Rating: 2
If anyone is intersted:

The real first production AIB ship is March 12, 2007. Embargo date is, and has been, the week ending March 30, 2007.


RE: No one noticed this yet?
By Griswold on 2/20/2007 7:03:48 AM , Rating: 3
You forgot to mention "OS Version: 7".

Last time I checked, XP was version 5 and Vista version 6. He must have this brandnew, to be released Windows Vista XP or something. :P


RE: No one noticed this yet?
By johnsonx on 2/20/2007 11:21:05 AM , Rating: 2
Actually XP is Windows NT v5.1; 2000 was NT 5.0. But you're right, Vista is version 6. There is no Windows version 7 yet. That might be Sausage, due out in 2009...er...2010...er...whenever...


RE: No one noticed this yet?
By johnsonx on 2/20/2007 11:44:18 AM , Rating: 2
Another little detail from the 'result' screen that jumps out at me is that it identifies the CPU as 'Family 15, Model 1, Stepping 3'. Family 15 is correct enough, and Stepping 3 certainly seems plausible. However, Model 1 doesn't look right to me. The old S754 Newcastle I'm typing this from is Model C. While I don't have any other Athlon64 boxes handy at this moment (and certainly no FX-74!), I don't recall ever seeing Model 1 for any K8 product.

Now this doesn't prove anything to be bogus (in it's Beta state it could just be reading the wrong register), but like many other details here it is odd.


Video
By KristopherKubicki (blog) on 2/20/2007 2:11:51 PM , Rating: 2
I highly suggest everyone take a look at the video on Fileshack. It's, uh, interesting.




RE: Video
By Ratwar on 2/20/2007 3:26:07 PM , Rating: 2
... You know, that video looks... Well... Let's just say I've seen more convincing videos from freshmen in a Java class...


RE: Video
By MrTeal on 2/20/2007 9:04:03 PM , Rating: 2
Weird. What's with the triangle of water in front of the left boat with no fancy twinkly lights?
Why don't the colored sphere cast light on the buildings?
Why is it so strangly jerky?

I know almost nothing about such things, but as a layman I have to say that such a video makes me less likely to believe this.


RE: Video
By nurbsenvi on 2/20/2007 10:24:33 PM , Rating: 2
My question is why is Fudo doing this?
Though he might be cold blooded cynic I thought he was at least kinda entertaining.

I can't help notice his bias towards ATI time to time (I think he has close friend working there) and now going so far as fabricating a none existing benchmark is just timid and stupid!

Any way, Kris why do you hate The Inq so much? I think they are entertaining if you take it with grain of salt.


RE: Video
By KristopherKubicki (blog) on 2/20/2007 10:36:11 PM , Rating: 2
I don't hate them at all. If Anand wrote the same rubbish Fudo did I would just as surely refute his claims as well.


RE: Video
By Le Québécois on 2/21/2007 1:18:22 AM , Rating: 2
quote:
If Anand wrote the same rubbish Fudo did I would just as surely refute his claims as well.


THAT would be entertaining!!!

But back on the video, I've seen it and must say that I'm not sure how to react. While the water effects are ok the models are pretty simple, the background doesn't seem to have any effect and except for the light and the waves on the water, nothing is moving...

Maybe someone with better knowledge than I have on:

quote:
* True 64-bit High Dynamic Range Lighting with Antialiasing
* Parallax Occlusion Mapping
* Soft Shadows
* Normal Mapping
* Soft Particles
* Full Scene Motion Blur
* Depth-of-Field
* Heat Haze
* Volumetric Explosion
* Realistic Water Physics


...could explain to me what is special with this benchmark. As far as I'm concern, it could use all the most advance effects in the world, if the end result is ugly, I don't see the point.


RE: Video
By johnsonx on 2/21/2007 11:42:13 AM , Rating: 2
oh dear... that is rather sad, isn't it? The 2001 music makes the whole thing pathetic, or funny depending on your mood going in. S3 should ask for their 50 cents back.

Why don't the lights affect the boat models or the buildings? While I have no idea how to do what is shown, I strongly suspect it's trivial for someone who does know. It kinda looks like some kid trying out all his tricks after reading 'Direct3D for Dummies'.


RE: Video
By GaryJohnson on 2/21/2007 1:23:02 PM , Rating: 2
That was Xtreme.


might as well make the 2nd offer $1,000,000
By johnsonx on 2/20/2007 12:18:10 PM , Rating: 3
quote:
I'll donate another $1,000 if it can display the images from the RyderMark "screenshots" published in July 2006.


KK,

You may as well make that second offer $1,000,000. Those screenies were fakes, EVERYONE knows it including Faud (and not just the sharpening tool thing... the entire image was a poor photochop). There may be a real RyderMark (sponsored by S3 though? lol!), but there's no way it could produce those exact images.

What kills me is that Faud STILL won't admit those original images were faked. He could simply explain that those images were mock-ups someone at Candella had done for what RyderMark was supposed to look like, and then there was miscommunication down the chain until someone thought those were actual screenshots. The truth may not be far off from that scenario, as I can't imagine anyone would intentionally try to pass those off as 3d rendered scenes.




RE: might as well make the 2nd offer $1,000,000
By AlabamaMan on 2/20/2007 11:12:53 PM , Rating: 2
The problem is, he can't exaplain them away as mockups, unless he is going to clam that Nvidia hardware causes image corruption in Photoshop.


By johnsonx on 2/21/2007 11:26:49 AM , Rating: 2
I suppose the explanation could be that the 'sharpened' image mockup was originally intended to show the sort of difference Candella expected between different shader precision settings, or something like that. After that, once the images were mistaken for, and believed to be, true screenshots, hilarity resulted.

I dunno, I'm not trying to explain the fiasco away at all. I'm simply pointing out that Fraud's.... er... I mean Faud's failure to address this issue at all is maddening to all concerned. We're all looking at a smoking gun, and Faud continues to pretend there's no gun at all much less any smoke... problem is there's so much smoke pouring from that gun, it's hard to see any truth at all about anything else Faud says or anything to do with RyderMark.


By James Holden on 2/21/2007 1:19:03 AM , Rating: 3
I feel cheated and lied to. I am out of work right now, so I have spent a considerable amount of time analyzing the "news" articles on Inquirer for R600. I've been waiting for this card for months, and it looks like I'll wait a little longer.

Thank you Inquirer for shamelessly pumping out rumor after rumor, a disservice to your readers and especially to me.

May 22, 2006
http://www.theinquirer.net/default.aspx?article=31...
The largest chip of all time
Hard launch before the end of the year

May 26, 2006
http://www.theinquirer.net/default.aspx?article=31...
1GHz clock speed
launch not for 2006
Might be a 65nm chip

May 30, 2006
http://www.theinquirer.net/default.aspx?article=32...
R600 will not need a Crossfire cable

April 4, 2006
http://www.theinquirer.net/default.aspx?article=30...
R600 delayed because Microsoft Vista is delayed
Early Q4 launch

June 13, 2006
http://www.theinquirer.net/default.aspx?article=32...
Hard Q4 2006 release

June 15, 2006
http://www.theinquirer.net/default.aspx?article=32...
R600 to launch with Microsoft Vista
64 unified shaders

June 21, 2006
http://www.theinquirer.net/default.aspx?article=32...
R600 taped out

June 29, 2006
http://www.theinquirer.net/default.aspx?article=32...
R600 has new cooler design

July 18, 2006
http://www.theinquirer.net/default.aspx?article=33...
R600 is 80nm

August 3, 2006
http://www.theinquirer.net/default.aspx?article=33...
R600 has a new cooler design

August 10, 2006
http://www.theinquirer.net/Default.aspx?article=33...
R600 is 80nm node

August 18, 2006
http://www.theinquirer.net/default.aspx?article=33...
R600 has 64 pipes
January launch date

September 14, 2006
http://www.theinquirer.net/default.aspx?article=34...
R600 launch moved from December to January

September 18, 2006
http://www.theinquirer.net/default.aspx?article=34...
R600 to consume 250W
Hard January/February launch

September 19, 2006
http://www.theinquirer.net/Default.aspx?article=34...
GDDR4 for R600 clocked at 1GHz

September 20, 2006
http://www.theinquirer.net/default.aspx?article=34...
R600 has clock higher than 700MHz
64 unified shaders

October 2, 2006
http://www.theinquirer.net/default.aspx?article=34...
R600 has 500 million transistors

October 13, 2006
http://www.theinquirer.net/?article=35062
R600 has 512Mbit memory bus
128GB/s memory bandwidth
February hard launch

October 19, 2006
http://www.theinquirer.net/default.aspx?article=35...
R600 higher 3dmark score than 8800 geforce

October 30, 2006
http://www.theinquirer.net/?article=35406
R600 die smaller than 480mm^2

November 1, 2006
http://www.theinquirer.net/default.aspx?article=35...
R600 has native HDMI

November 8, 2006
http://www.theinquirer.net/default.aspx?article=35...
R600 scheduled for Q1'07 launch
Silicon released

November 15, 2006
http://www.theinquirer.net/default.aspx?article=35...
16 pixels per clock

November 15, 2006
http://www.theinquirer.net/default.aspx?article=35...
12 layer PCB
Memory bandwidth between 115gb/s and 140gb/s
GDDR3 or GDDR4

November 15, 2006
http://www.theinquirer.net/?article=35752
Late january launch optimistic

November 28, 2006
http://www.theinquirer.net/default.aspx?article=35...
R600 has more transistors than g80
500+ GFLOPS

December 1, 2006
http://www.theinquirer.net/default.aspx?article=36...
February launch
Coincide with Microsoft Vista

December 1, 2006
http://www.theinquirer.net/default.aspx?article=36...
R600 is GDDR4 only
1GB and 2GB versions
Early March launch

December 6, 2006
http://www.theinquirer.net/default.aspx?article=36...
R600 same time as Vista

January 2, 2007
http://www.theinquirer.net/default.aspx?article=36...
R600 driver released
Hard February launch

January 7, 2007
http://www.theinquirer.net/default.aspx?article=36...
R600 demonstrated
1GB of GDDR3

January 9, 2007
http://www.theinquirer.net/default.aspx?article=36...
Late february, early March launch
Silicon is done, again

January 17, 2006
http://www.theinquirer.net/default.aspx?article=37...
R600 uses air cooler
launch in time for CeBIT

January 18, 2007
http://www.theinquirer.net/default.aspx?article=37...
Dual HDMI confirmed
Four variants at launch: R600 XTX, XT, XL, GT

January 24, 2007
http://www.theinquirer.net/default.aspx?article=37...
R600 final silicon done, again
Firm march launch date

January 28, 2007
http://www.theinquirer.net/default.aspx?article=37...
R600 "on track" for Q1 07 launch,
Which we say as March including availability

February 2, 2007
http://www.theinquirer.net/default.aspx?article=37...
Early march launch
No GDDR3 version

February 2, 2007
http://www.theinquirer.net/default.aspx?article=37...
R600 to launch in early march

January 3, 2007
http://www.theinquirer.net/default.aspx?article=36...
Early march launch
720 million transitors

February 6, 2007
http://www.theinquirer.net/default.aspx?article=37...
March r600 launch

February 6, 2007
http://www.theinquirer.net/default.aspx?article=37...
Four flavors of R600: xtx, xt, pro, gt

February 7, 2007
http://www.theinquirer.net/default.aspx?article=37...
Quad R600
Radeon x2800 name

February 9, 2007
http://www.theinquirer.net/default.aspx?article=37...
R600 to use samsung SGRAM
179.2GB/s memoryu bandwidth

February 9, 2007
http://www.theinquirer.net/default.aspx?article=37...
4GB and 2GB versions of r600

February 9, 2007
http://www.theinquirer.net/default.aspx?article=37...
R600 launch in early March

February 9, 2007
http://www.theinquirer.net/default.aspx?article=37...
R600 water cooler
DDR3 and DDR4 models

February 12, 2007

http://www.theinquirer.net/default.aspx?article=37...
Four launch versions of R600

February 13, 2007
http://www.theinquirer.net/default.aspx?article=37...
External R600
March and April launch

Was originally published Feburary 19, 2007, dated changed to February 14, 2007
http://www.theinquirer.net/default.aspx?article=37...
The entire article republishes specifications posted on Dailytech on February 19, 2007
Contradicts articles published on the same day

February 14, 2007
http://www.theinquirer.net/default.aspx?article=37...
X2800XTX has 750MHz clock

Feburary 14, 2007
http://www.theinquirer.net/default.aspx?article=37...
Radeon X2800XTZ2 $600
Dual GPU design

February 16, 2007
http://www.theinquirer.net/Default.aspx?article=37...
Radeon X2800XTX has watercooling

February 19, 2007
http://www.theinquirer.net/default.aspx?article=37...
Late March launch
Might be called X2800XTX or X2900XTX

February 19, 2007
http://www.theinquirer.net/default.aspx?article=37...
HDMI on r600, Radeon X2800




By oTAL (blog) on 2/21/2007 7:01:13 PM , Rating: 2
It really is true... when one guesses on all the possible outcomes, it is pretty easy to "predict" the future. That's how they managed to be right on some (few) of those articles.
Makes me remember a Portuguese "seer", who named himself Zandinga, who predicted in his "new year predictions" that the pope would die... and he did...
Two things come to mind...
1. The pope was a pretty old guy and I can pretty much guarantee that the current pope won't make it past 2020... that doesn't make me special or empowered with mystical supernatural cosmic metachloridians.
2. Lets go back a year and check his "predicitons"... oh! "the pope is going to die!"... now I'm kinda *predicting* that if I go back another year... yup! there it is again... powerful this guy, huh?

Inquirer FTW! =p


Come on now!
By Beany on 2/19/2007 5:47:09 PM , Rating: 2
I seriously cannot believe people cant instantly tell that those two original shots (one is posted above in the article), apparently running on ATI and nVidia hardware are quick Photoshops. It's SO obvious.
Ok i've been using Photoshop for over a decade, but still, it's so blatant they are not screen captures of the benchmark running. I could go on about the images, how you can see the sharpening, erasing, blurring, brightening, cropped images and so on... but the point is i shouldn't even need to because it's so obvious it's a photoshop the second i see it.
It's like when games developers often release early shots of a game, and those shots are obviously a CG rendered image, made in 3DSmax/Maya/Cinema4D or whatever, and are not actual in-game shots (funny how this mostly happens with new Sony consoles). Yet so many people fall for it, still.

It honestly depresses me how thick people are.




RE: Come on now!
By johnsonx on 2/20/2007 11:48:29 AM , Rating: 2
Which people are you referring too? AFAIK EVERYONE knows those images are photochops, and piss-poor ones at that. The only thick one is Faud for not admitting the original screenshots were bogus. There may actually be a real Rydermark now, but him continuing to stand by those original screenshots just casts doubt over the whole thing.


Deadline
By SSH on 3/1/2007 6:32:46 AM , Rating: 2
Hang on, you gave them 10 and a half hours to respond?




RE: Deadline
By James Holden on 3/5/2007 11:37:37 PM , Rating: 2
And they did. And they declined.


<no subject>
By Scabies on 2/19/2007 4:45:30 PM , Rating: 2
...I never heard of this story, but after reading their post, I cant help but hear it as "omfg I am the 1337xxorest-est!"
And those weird seams... Crossfire? Wait he said NVidia...
(DT, this kid might be more inclined to share his 1337ware with you if the $2k goes directly to him.)




More photochopping
By Facey on 2/19/2007 11:20:41 PM , Rating: 2
I have only looked at the second of his new screenshots, but I am pretty sure it has been photoshopped: http://www.modtheater.com/forum/showthread.php?p=2...

I think your money is safe. :)




Candella Software
By GaryJohnson on 2/21/2007 5:05:44 PM , Rating: 2
I get the impression that:
-some of the games their site claims they have worked on they might not have.
-some of the companies their site claims are partners may not be.
-a lot of the 3D artwork on their sites has been manipulated in photoshop.
-their contact address might be an apartment in London.
-the whole company might only be one person: Ajith Ramachandran
-whoever does their art likes horses and speedboats.

Their site:
http://www.candellasoftware.com/

The site for their game "Chariots":
http://www.planetchariots.com/

An interesting old E-Bay listing for some land for sale by Candella:
http://cgi.ebay.co.uk/TROPICAL-LAND-NEAR-BEACH-KER...

Someone named Ajith Ramachandran is or was working for S3:
http://www.google.com/search?q=%22Ajith+Ramachandr...

A site with more info (including a personel count) on Candella:
http://www.gamesoutsourcer.com/companies.php?id=78

Their phone number (in London?):
+44 (0) 208 445 4197




Bye Bye Fudo
By CyborgTMT on 2/23/2007 6:26:32 AM , Rating: 1
Looks like Fuad decided to pick up his ball and go home...

http://www.theinquirer.net/default.aspx?article=37...

IMO the INQ's credibility just went up 30%




RE: Bye Bye Fudo
By Belegost on 2/23/2007 10:11:06 AM , Rating: 1
1.3*0 = 0


"Nowadays, security guys break the Mac every single day. Every single day, they come out with a total exploit, your machine can be taken over totally. I dare anybody to do that once a month on the Windows machine." -- Bill Gates













botimage
Copyright 2009 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki