Author: Robert Hyatt
Date: 15:26:57 12/02/02
Go up one level in this thread
On December 02, 2002 at 17:21:07, Bob Durrett wrote: >On December 02, 2002 at 13:19:16, Robert Hyatt wrote: > >>On December 02, 2002 at 11:48:26, Bob Durrett wrote: >> >>>On December 01, 2002 at 23:02:44, Robert Hyatt wrote: >>> >>><snip> >>> >>>>Only "children" pay more attention to the pictures than to the words written, >>>>so most of us wouldn't consider that a big deal. <snip> >>> >>>But Bob! I LIKE pictures. Does this mean that I have already slipped into my >>>"second childhood"? MUST I read all those words in my National Geographic >>>magazines? >> >>No. But it means that with a _technical article_ about a product that is yet to >>be >>released, a photo is meaningless. The text of the article is the "meat" and >>describes >>the important features and performance. The photos are "window dressing". >> >> >>> >>>Besides, there are many highly respectable professionals who devote their entire >>>lives to the creation and editing of pictures. The use of software is very >>>common and definitely is a "big deal." >> >> >>No argument. But we are talking about a long, technical discussion about a new >>processor. What technical content is there in a top view of the CPU, other than >>to >>say "see, it exists, here is a picture." There are no specifications or >>anything other >>than the part number that will provide the clock rate... and the article >>already has >>that in the first paragraph... >> >> >> >> >> >>> >>>Also, someone who examines a picture to determine how the picture was put >>>together, and to discern the picture's information content, should be respected. >>> Would you get rid of all "photo" analysts in the intelligence services? >>> >>>Stand corrected? : ) >> >> >>No, because we are talking in two different worlds. Do you ever read the "spy >>photo" >>deals in a magazine like Popular Mechanics, where they show _next_ year's >>brand-X >>car? You read the article to find out about the drivetrain, etc, you look at >>the photo >>to get a rough idea of appearance if appearance is important to you. Otherwise >>you >>know that the photo is of a prototype anyway and that the final product will >>look >>different. But the known _details_ are important. >> >>That happens in electronics as well. You send me a chip, say "you can have it >>for >>24 hours" but it must be fed-exed to person X by tomorrow at 3pm. You start >>testing >>and when you go to snap a digital photo of the chip, your camera is dead. Do >>you spend >>the rest of the day working on getting a pretty-well useless photo, or do you >>drop the >>photo issue and test like crazy to learn as much as you can for your review? >> >>I know what I would do there. I might even resort to a doctored photo although >>I would >>certainly say "this photo is composed to show how the product will look when it >>is delivered, >>a real photo was unavailable at press time." or something similar. >> >>Again, the photo was not the news. The details about the chip was the important >>thing. > >O.K. I'm convinced. I'll start reading the text and quit looking at those >pictures. But . . . I hate that! I always believed "A picture is worth a >thousand words." You are destroying my faith in old sayings. : ( > >Anyway, I agreed with you all along. Just trying to get you to laugh a little, >Bob H. : ) > >On a more serious note, if someone is reading an article with the intent to get >the most out of it, that person should look at everything offered. At least, >that makes sense to me. Rejecting the text because of something odd in the >photo seems counterproductive. But ignoring the photo seems unwise as well. I would never say "ignore the photo". I would say take the entire article in context and if there appears to be an accurate text review, with a photo that seems "odd" then so what? In this case there are several plausible explanations for a doctored-up CPU photo. Any could be right. All could be wrong... > >One other thing needs saying: Nothing is true just because it is written down. >Pictures can lie too! The reviewer may or may not have presented fully factual >information. A degree of skepticism seems prudent when reading reviews, >especially if the product being reviewed is not yet available. The reviewer, >being human, may have a hidden agenda! He could be prejudiced. Or, maybe, he's >getting a kickback "under the table" from someone? It happens! > I don't disagree. However, I generally take things at face value if they are from a reputable source. And I generally ignore them if they are not (the comment about Leko from Sam Sloan is an example of the latter case). THG has always seemed legit to me, and AMD provides them with information and pre-production hardware, which means AMD must think they do reasonable reviews or they would "pull the plug" on them and not give them stuff to review prior to it being available to the public. That at least suggests (to me) that THG is reasonably accurate. Do they make mistakes? They _must_ as everyone does. But does a mistake make them dishonest? I don't believe so... >Blind faith in the goodness and honesty of people everywhere has a lot of merit >IMHO and does have it's place. But maybe not here? If you will permit me >another "old saying," then try this one: "All is fair in love and war." Some >people love AMD and others love Intel. And . . . "business is war." True, but THG gets hardware from AMD, Intel, even HP and others. If they are too biased some of those sources would dry up quickly... > >Incidentally, I doubt that either AMD or Intel care one whit about chess >engines. You'd be wrong, because Crafty is in _the_ benchmark that everyone uses to compare raw CPU power. I've talked to _every_ chip manufacturer over time, helping them understand the code, helping them track down compiler bugs with high levels of optimization, etc. They aren't paranoid about it, but they do take notice since chess is "in the mix." However, THG is not about chess anyway, it is about hardware information and comparisons to show what is what and when it is/isn't... > >Peace. [It's that time of year.] > >Bob D. > > >> >> >> >> >> >> >>> >>>Bob D.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.