SDRAM vs. RDRAM, Facts and Fantasy - Page 2
Home | Reviews and Features | Special Reports | Forums |

Page 2 of 6 FirstFirst 1234 ... LastLast
Results 16 to 30 of 84

Thread: SDRAM vs. RDRAM, Facts and Fantasy

  1. #16
    Join Date
    Sep 1999
    Location
    Netherlands, Europe
    Posts
    170
    Hello there,

    I'm really suprised that most of you are suggesting that I or even HWC is 'owned' by either Rambus or Intel. Allow me to fill you in on some of the work that has gone into writing this article. Most of the info you'll see posted came from Rambus, Intel, Samsung, Micron, VIA and AMD a whole slew of company datasheets and a compilation of other websites' data.

    I've asked each and every one of these manufacturers to send me a package with some indepth technical background info, which I them compiled into the info you see posted. I think I've done a fair comparison by all means, all the drivers and benchmarks used were recommended to me by the manufacturers of the applicable chipset/motherboard/manufacturer, you can be sure that they make sure I use a driver/benchmark that has their product performing at best.

    One of the things that really offends me with these kind of flames is the fact that you people tend to absorb all anti-Rambus/Intel articles as the word of God, but flame one that does shed a different light and dares to take a different point of view. Who's being objective here? I disclose all info, drivers used, benchmarks etc. I don't see any of that in those other 'articles', they just summarize their results without posting a single screenshot and you tell me you're buying that?

    I don't own any Intel or Rambus stock, nor any stock for that matter. I don't get company kickbacks, or get whatever CPU + memory I want for personal pleasure for free. I think you should think twice before you start making these kind of claims.

    I hold a masters degree in EE and a bachelors in ME, so you can be sure I'm able to tackle any technical issues, thus I might not do the benchmarks you'd like to see, I do however make sure they're objective and as well documented as possible, the same cannot be said about many others.

    I'm all for constructive critisism, but come up with valid arguments and support them with a link, a quote, etc, don't just state that I did a bad job, as that is soo easy. To quote some of your own statements; 'If you want to play with the big boys in this arena, you better be more thorough and fair-handed than you have with this poor showing. Shame, really.'

    Thanks and best regards,

    Sander Sassen
    Siteleader at HardwareCentral
    Email: ssassen@hardwarecentral.com
    Visit us at : http://www.hardwarecentral.com


  2. #17
    First of all, it IS easy to flame, but then I'd say, when you flame do so in style. (Do it with facts).
    Personally I welcome this article, it shows RDRAM in another light. But as others allready pointed out, it is clear in pointing out the merits of RDRam versus the 'bad' things in SDRam. And I am missing the 'bad' things of RDRam, cauz when it's that good, why does it fail to outperform SDRam ?

    My it is because your used sources did hide this information from you ?

    Also I'd rather see a comparison of RDRam vs. DDRam. Since there almost 'equal' in age and and have a similair bandwidth. Micron itself has allready stated that allthough the bandwidth of the newest types of DDRam is higher, due to it's less efficient protocol it approaches that of RDRam. (Sorry no link for support)

    Then again (dunno where I read it but) this might be one of those attempts of Intel to seize control over the (pc)world again.

    For a personal note...I keep saying this is a smelly kinda memory.

    ------------------
    driesens.com

  3. #18
    First of all, it IS easy to flame, but then I'd say, when you flame do so in style. (Do it with facts).
    Personally I welcome this article, it shows RDRAM in another light. But as others allready pointed out, it is clear in pointing out the merits of RDRam versus the 'bad' things in SDRam. And I am missing the 'bad' things of RDRam, cauz when it's that good, why does it fail to outperform SDRam ?

    My it is because your used sources did hide this information from you ?

    Also I'd rather see a comparison of RDRam vs. DDRam. Since there almost 'equal' in age and and have a similair bandwidth. Micron itself has allready stated that allthough the bandwidth of the newest types of DDRam is higher, due to it's less efficient protocol it approaches that of RDRam. (Sorry no link for support)

    Then again (dunno where I read it but) this might be one of those attempts of Intel to seize control over the (pc)world again.

    For a personal note...I keep saying this is a smelly kinda memory.

    ------------------
    driesens.com

  4. #19
    Join Date
    Feb 2000
    Location
    Deutschland
    Posts
    936
    I don't know, Sander... Claiming that everyone else doesn't give the details is a bit of a stretch.

    Granted, some sites do an even poorer job at benchmarking. And in all fairness, this hapens on both sides of the debate. I won't deny that. But then, I won't consider them to be professional, either.

    On the other hand, there are enough sites out there that do a very thorough job about it. As in: running a lot more benchmarks, under NT, Windows 98 and Windows 2000, and experimenting with the drivers to see what happens.

    It's precisely this second kind of test I'm interested in. Precisely BECAUSE it's a lot of work, and I really don't have the time (nor the funds) to do it myself. Frankly, if I wanted to settle for Sandra and Quake, with whatever OS and drivers happen to be installed, I'd just go to a friend's house and run those myself.

    But what I'm really interested in os that someone did a helluva lot of work tweaking and testing, so I don't have to do that myself.

    ------------------
    Moraelin -- the proud member of the Idiots' Guild
    Moraelin -- the proud member of the Idiots' Guild

  5. #20
    Join Date
    Apr 2000
    Location
    Wallaceburg Ontario Canada
    Posts
    16
    Moraelin et al:

    To be fair to Sander: I've not seen anyone do a comparison of DDR-SDRAM to RDRAM anywhere online yet - tho in the same breath, I will say that every other site thats done RDRAM vs SDRAM comparisons does mention the DDR-SDRAM in depth, and I've not seen that in either of the 2 articles that Sander has posted. I can excuse that I suppose by giving him the benefit of the doubt on that he is looking at current technology and whats available now.. not on whats going to be available. However, i do find it interesting that this is one of the few sites out there (actually.. the only one I've seen) that continues to insist that RDRAM is the "2nd Coming" (other then Rambus of course)

    More interesting in light that I've seen news articles in recent days that a couple of memory manufacturers are cutting back on RDRAM production because of poor yields and because most anticipate not having a very big market for RDRAM - focusing instead on the perceived more lucrative SDRAM market (or DDR-SDRAM market). There are also hints that Intel may be backing away from RDRAM and is starting to make sure a lot of its platforms are SDRAM compatible. So Sander, while I admire you and your site for trying to educate the masses, I'd say your conjecture that RDRAM is the next best thing to happen to the PC world is debatable at best.

  6. #21
    Join Date
    Jan 1999
    Location
    MI
    Posts
    4,144
    Stribe,
    eloquently put. I concur.

    Sander,
    I will post later with my debate.

    ------------------
    With greater knowledge comes greater understanding!
    With greater knowledge comes greater understanding!

  7. #22
    Join Date
    May 2000
    Location
    New York, NY
    Posts
    2
    Thank you for an informative article. If one wishes to consider it an article advocating a point of view, I wish all advocacy articles would be as well done.
    There is nothing wrong with that.

    However, these decisions are not always or even often decided on technical merit. The PC industry has always been based on affordable technology, and until RDRAM comes within shouting range of its competition, it will be stillborn or at best a niche product.

    Eventually, something like RDRAM will be needed, because, if nothing else, broadband video will eventually demand it. This does not mean RDRAM will inevitably become the next memory standard. There does not seem to be any pressing technical need for bandwidth on the average computer that DDR could not meet for the next generation. Afterwards, it is possible that the Intel consortium will come up with something that will leapfrog RDRAM.

    Often, huge technical decisions are made for personal reasons. We might be using CP/M 2000 if some IBM people didn't get peeved Gary Kildall vapored at a planned meeting. Given this, the general dislike of Rambus (hardly ameliorated by the current lawsuit) may outweigh any technical considerations. On the other hand, Intel's hamhanded pushing of RDRAM is hardly bereft of personal motivations, either.

    I am only saying it is not true that RDRAM must win. I am not at all saying RDRAM cannot win. If there is enough supply to feed at least Willamette when it ramps up, at a price reasonably close to DDR, RDRAM stands a good chance of winning. Then you could easily have the memory manufacturers jumping on the bandwagon throughout 2001. But RDRAM must achieve that by early next year. As a rough estimate, you'd have to have about 3 million 128Mb equivalent RIMM (not chip) capacity a month early next year to do that.

    Again, thank you for a very good article.

  8. #23
    Join Date
    May 2000
    Location
    Sydney
    Posts
    4
    i think wat i want to say is RDRAM cost so much more than the SDRAM and yet not even performing faster than SDRAM at all in most of the applications and games

    costing about 5x more than SDRAM, shouldn't it kicks in ALL test.

    however it doesn't seem to be the case. It is not even winning in the Q3 benchmarks. Dont we expect it to be so much faster when we pay about 5x more? we dont want it to be 'slightly' or 'sometimes' faster.

    Also, comparing the BX at 100, VIA at 133 and i820 at 133 is not fair for the bx

    also, testing the Best quality RDRAM, (PC800) but then comparing it with a VIA PC133 (CAS3) is simply not fair dont u think?
    Cas 2 is about 2-3% faster than cas3 in all cases and since the best RDRAM is used, the best SDRAM should be used as well. VIA is KNOWN to be slower than the BX.

    basically you are comparing i820 at 133 compared to BX at 100MHz FSB. and now you say it is a 'fair' comparison. Wat do you think?
    isn't it ashame of rambus and intel that i820 even at 133/RDRAM is about the same performance of a BX at 100/SDRAM...


    also, using a Soyo BX motherboard is another thing i wana say

    you have used an Asus P3C-E. Why dont u use an Asus P3B-F as well to test the SDRAM? another example of how 'fair' you are

  9. #24
    Join Date
    May 2000
    Posts
    1
    No technoeconomic analysis is complete if it does not cover threat perceptions. In management parlance it is called SWOT analysis (Strength,Weakness,Oppurtunities,Threat). If you are talking about present, then the price definitely holds it down. If you talk about future, threat(DDR) is REAL and needs to be discussed (market window). No technology decision or investment decision can be accurate if only based on current technical advantage or price advantage.

    ------------------
    T.Sridhar

  10. #25
    Join Date
    Feb 2000
    Location
    Deutschland
    Posts
    936
    Actually, Stribe, I've seen at least one place benchmarking the Micron Samurai DDR chipset with ONE channel DDR, versus the i840 with TWO channels RDRAM. It spanked the dual channel i840 in all tests. (In some of them marginally, but in some it spanked it real badly.)

    In fact, Sander himself pointed us at that article. (And called it something ignorant and ill educated people will use in their grudge against Rambus. As you can see, flaming around here goes both ways )

    I don't remember the link, but if you'll look through Sander's recent articles, I'm sure you'll find it.

    But, again, I'll admit we must thank Sander for including some real world benchmarks this time. Flawed or not, at least he did try. And that driver mix-up is probably an honest mistake. After all, most of us would have went with the newest drivers, too. (And, heck, if someone is to blame about it, is VIA, who still can't get their act together and make good drivers. Remind me why I avoid their boards.)

    ------------------
    Moraelin -- the proud member of the Idiots' Guild
    Moraelin -- the proud member of the Idiots' Guild

  11. #26
    Join Date
    May 2000
    Location
    Jacksonville, FL, USA
    Posts
    3
    RDRAM Pet Peeve:
    A recent article in PC World states that the price premium OEMs pay for RIMMs is much lower than the premiums in the retail channels, suggesting that the inherent costs of RDRAM are much lower than initially reported.
    Not necessarily. That is only one possibility. Rambus Inc. might have elected to take a short-term loss with large OEMs in order to get the technology into the market and acquire marketshare thus helping acquire mindshare. Then, hypothetically, when Rambus-Intel managed to make RDRAM dominant, they can then up the price to recoup losses. Furthermore, alot of the parent PCWorld article relies (much like Sasser admittedly did) on the marketingspeak of the PR departments of parties involved in the success of RDRAM. Never look a gift horse in the mouth! Rambus is quite fond of quoting Dell and HP systems, but on further price analysis, you can see just how much more you are spending on RDRAM systems for poor performance differences.

    Also, yields seem to be very low. Therefore, my guess is that they can barely fufill OEM orders, leaving a paltry supply for the retail channels. Yes, this will inflate price and that's probably why it costs an extreme 5-7x price difference, but until yields and supply increase, this won't change and I feel it won't happen soon. (At least not before DDR kills its last chance to get mindshare.)

    - alphadog

  12. #27
    Join Date
    Apr 2000
    Location
    Philadelphia PA
    Posts
    7
    Dear Sander,

    Sorry to hear that you are stung by our questions of your capabilities. But it seems to us you have taken up a crusade to defend Rambus. DON'T JUST TALK TO RAMBUS EMPLOYEES! Memory technology is extremly complex and you need to be a specialist in that area to give conclusions you are propounding. I don't pretend to be. Don't be that upset when Rambus in PC architecture dies a slow death. A lot of people have fallen for Rambus' PR. Rambus is a business venture. They care more about their stock price than technology.
    I won't even talk about their shady patent ammendments based on freely given information.
    Looking at Rambus' site I see similar lines of arguments as yours. The claim about no latency problem for Rambus (from Rambus" is classic "double-think".
    I also feel your past and present articles are spoiled by leaps to conclusions that are unreasonable (not just this article).
    Your benchmarks are inconclusive, yet you say Rambus is a clear victor. One can look at these results and go either way with a conclusion depending of one's inclination or favoritism.
    Much of what you say is based on faith that the scores will improve when support is improved and processor speed has improved. Of course, this may actually happen. But it is not backed up by the benchmarks.
    BTW who recommended the synthetic benchmark Sandra as a memory test?
    How does Sandra actually measure memory performance? What does the Memory Industry use? I know they don't use "Sandra"! Did you ask Micron what they used?
    You again refer to criticism of Rambus as "Fantasy". Yet you persist in claiming Low Latency for Rambus that is plainly wrong! Who is suffering from "Fantasy"? Your arrogant remarks about "Tom's Hardware" will come back to haunt you someday.
    This is why doubts about your views are expressed. And frankly, as a source of computer information, I will in the future go elsewhere and not waste my time here.
    Adios

  13. #28
    Join Date
    May 2000
    Location
    Copenhagen, Denmark
    Posts
    1
    [QUOTE]Originally posted by Sander Sassen:
    [B]Hello there,

    I'm really suprised that most of you are suggesting that I or even HWC is 'owned' by either Rambus or Intel. Allow me to fill you in on some of the work that has gone into writing this article. Most of the info you'll see posted came from Rambus, Intel, Samsung, Micron, VIA and AMD a whole slew of company datasheets and a compilation of other websites' data.

    I've asked each and every one of these manufacturers to send me a package with some indepth technical background info, which I them compiled into the info you see posted. I think I've done a fair comparison by all means,
    - No, not by all means!

    One of the things that really offends me with these kind of flames is the fact that you people tend to absorb all anti-Rambus/Intel articles as the word of God, but flame one that does shed a different light and dares to take a different point of view.
    - That you dare, does not make your points more correct!

    Who's being objective here? I disclose all info, drivers used, benchmarks etc. I don't see any of that in those other 'articles', they just summarize their results without posting a single screenshot and you tell me you're buying that?
    - For the articles I have read, - yes I'm buying it!

    I hold a masters degree in EE and a bachelors in ME, so you can be sure I'm able to tackle any technical issues, thus I might not do the benchmarks you'd like to see, I do however make sure they're objective and as well documented as possible, the same cannot be said about many others.
    - You might be able to tackle technical issues (although I still don't think you're doing a fair comparison from a technical standpoint), but you lack in marketing.

    I'm all for constructive critisism, but come up with valid arguments and support them with a link, a quote, etc, don't just state that I did a bad job, as that is soo easy.
    - Okay, here we go:

    One the things that offend me most is the question about price. In the article you write in the chapter about RDRAM pricing
    that, and I quote: "Price is undoubtedly the most controversial aspect of RDRAMs today.", and by this statement you are making the price a central point about RDRAM. Further down in the same sentence you give the impression that RDRAM in fact are not that expensive by reffering to Dell's website. As you hold a masters degree in EE and a bachelors in ME, I guess it couldn't be diffecult for you to find out about what kind of RDRAM they were talking about, because Dell nicely provide specification for it! Namely PC600 RDRAM (actually going at 266 MHz, which only provide the same peek bandtwidth as PC133 RAM). Still you unfairly benchmark a PC800 RDRAM, which is much more expensive, not to say almost impossible to get (whether in retail or OEM)!
    If you really want to remove this feeling that you are bought by RAMBUS or Intel, you should make a benchmark based on this RDRAM and make your conclusions thereafter!


    Sander Sassen
    Siteleader at HardwareCentral
    Email: ssassen@hardwarecentral.com
    Visit us at : http://www.hardwarecentral.com


    I'm a sys-admin who are buying quite a lot of computers for the university where i'm currently employed, but are extremely frustrated about the current situation in the supply and configuration of PC's. I guess this year will be a relaxed one regarding upgrading mashines!

    CH @ che@cph.ih.dk


  14. #29
    Join Date
    Apr 2000
    Location
    Wallaceburg Ontario Canada
    Posts
    16
    With regards to the charges of bias of this site towards Rambus/Intel,and Sander's vigorous denial of it, I can see from a certain standpoint why some people might think this the case:

    -- Most of the advertising you see on here appears to be either Dell or Intel based computers - obviously who have agendas with regards to Rambus...
    -- Hardware Central posts their April Review for computers which does not pick a single AMD based Athlon machine as a buyers choice, instead going Pentium III across the board - one of the only sites to exclusively pick Pentiums - as every other site thats to do with hardware who've had a buyer review out had a mix of both Athlons and P III's depending on speed and features.
    -- Hardware Central does a nice article on the Williamette.. but I've not seen any such articles on any of the AMD models coming out; those being the Duron, The Thunderbird enabled Athlon, or the Corvette/Mustang, OR next years 64 bit CPU SledgeHammer.
    Add to that Sander for whatever else you can say about him as at least been consistent in his praise of RAMBUS and RDRAM stretching back several months in many of the articles I've seen him post.

    So.. I can see why charges of bias erupt against this site occasionally, because I personally would like to see some more AMD coverage myself, being partial to that company and their line of CPU's.. but folks.. for a really good article on RDRAM vs (DDR) SDRAM, you need look no further then this very site - Posted Feb 20, 2000 by HWC's Dan Mepham. I found this article to be very objective looking at both options, and mayybe, just maybe in slight favour of DDR SDRAM.. I espescially like this quote:

    "At present, DDR SDRAM appears to be the more viable of the two solutions. DDR SDRAM won’t require a large investment on the part of
    manufacturers, and has thus far experienced decent yields. DDR simply seems, at present, to be the easier of the two options, while still offering performance on par with RDRAM.
    For the time being, neither is an excellent option. We have not yet seen a DDR chipset release, and RDRAM, while available, is far too costly for most users. If you’re looking to build a system within the next month or two at least, standard SDRAM will likely be your only option."

    His full story can be found at this URL: http://www.hardwarecentral.com/hardw...al/print/1519/

    So folks.. disagree with Sander all you want, but dont think that everyone at HWC is all ga-ga over RDRAM.. unless Dan has changed his mind in the last 3 months, theres divergent opinion it appears here as well.. It would be interesting to hear from Dan and what his thoughts are on Sander's articles - (tho I know its not always kocur to criticize a fellow columnist if he doesnt agree with it)

    Apologies for the long winded post.

  15. #30
    Join Date
    May 2000
    Location
    mtl
    Posts
    1
    Thanks Sander for the interesting article;
    it seems like i've missed some nice info on Rambus technology.

    Unfortunaly, it also seems like all the advantages of Rambus mentionned in the article are an transcript of Avo Kanadjian (VP of worldwide marketing at Rambus Inc...)answers in his interview at Firingsquad ( http://www.firingsquad.com )

    So is it me or did Rambus prodided you with all the findings in your 'in-depth investigation'


    Btw: I don't think that there's nobody worse than Dell when it comes to being bias toward INTEL

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •