Comments Locked

46 Comments

Back to Article

  • reload1992 - Wednesday, February 27, 2008 - link

    I am building my first pc. Is this a good cpu for gaming
  • Nil Einne - Monday, June 11, 2007 - link

    The reviewer correctly pointed out at the beginning that AMD has remained competitive by aggresively pricing their processors. Then at the end, they correctly pointed out that for non-overclockers, the BE-2350 isn't a bad choice. But the reviewer then somewhat stupidly suggested that people wanting a quite office system should just go with the E4300 because it offers better performance. This misses several points. Firstly if you just have an office system, you're probably rarely using the extra performance so it's questionable whether it's worth it. Secondly, the reviewer didn't consider mobo price. From what I've seen equivalently speced AMD mobos tend to be a bit cheaper then Intel ones.

    More importantly perhaps, the reviwer didn't consider other Intel or AMD offerings. There are the Pentium Es as others have mentioned. But there are also the older AMD chips. The X2 4200+ EE for example if you're looking for a low power chipped. And if you're not quite so worried about power, the X2 4600+ is about the same price as the E4300 but generally performs better. For a typical office system, the processor will mostly be in CnQ mode anyway so the actual power consumption probably won't change much.

    So unfortunately, it appears the reviewer, despite agreeing at first that AMD is competitive, has then completely missed the point of why AMD is competitive. AMD may be clock for clock slower, but they've priced their processors aggresively enough that they remain competitive for non overclockers from a price/performance standpoint. This trap is something a lot of reviwers seem to fall in sadly. Rather then saying stupid things like well the Intel system is faster and only a 'little bit' more (even when you don't need the extra performance) so you should go Intel, what they should be acknowledging is that price/performance wise they're fairly competitive because you can (at the low to medium end) get a better performing AMD system for a 'little bit' more too. If editors aren't going to consider the details, they should just refrain from making frankly silly comments in the conclusion.
  • psychobriggsy - Wednesday, June 6, 2007 - link

    AMD will be positioning these chips to be used with the AMD690G chipset.

    The Tech Report found that the nVidia SLI chipset used 14W more than the 690G (when used with an external graphics card).

    Comparing SLI with the Intel P965 is rather disingenous, and makes the power consumption page worthless, and quite misleading. Why didn't you use nVidia's Intel SLI chipset to make things more even? Or at least have a test with a motherboard using AMD's chipsets?
  • cocoviper - Thursday, June 7, 2007 - link

    I agree wholeheartedly.

    Anand could we get an update to the article testing equivalent chipsets? (Either an integrated graphics chipset on the AMD side vs. the 965; or retest the Intel power consumption numbers with an Nvidia SLI chipset?)

    Otherwise we might as well just toss out that entire power consumption page, because as it is right now the power delta between different mobo chipsets is as much or more than the power delta between the processors on that graph.
  • strikeback03 - Friday, June 8, 2007 - link

    quote:

    AMD argues that with integrated graphics you can deliver a lower overall TDP on one of its systems, which we tend to believe given that the P965 is one of the lowest power Intel chipsets you can buy whereas the nForce 590 SLI that's a part of our normal AMD test bed is far from the lowest power offering out there. We will be looking at integrated graphics/platform performance in a future CPU article when we will address this topic in greater detail.


    First paragraph, last page.
  • Nil Einne - Monday, June 11, 2007 - link

    I think you're missing the point tho. It was a bit silly to include the power graph in the first place since it's meaningless. Some reviewers seem to like to include graphs and data which is pointless at best and at worst misleading. This isn't exclusive to this instance or to Anandtech. There are a lot of stupid stuff out there (other examples include benching medium end cards at 1600x1200 with 4xAA and 8x Aniso). Other then wasting a readers time, the trouble here is that a lot of people don't properly read reviews, just skip through them and look at the graphs briefly. They see the graph and conclude the Intel has a lower power usage. They don't notice that the reviewer was bizarrely enough, comparing a SLI mobo with an integrated one so the graph is somewhat useless.
  • yacoub - Wednesday, June 6, 2007 - link

    My E4400 runs at 3.0GHz w/o breaking a sweat. $135 CPU that clocks to a $300 CPU's speed. I haven't been able to buy a CPU that cheaply literally in a decade. Last one that offered such a high ROI with a simple 50% overclock was the Celeron 300A (300MHz, ran at 450MHz).
  • myterrybear - Tuesday, June 5, 2007 - link

    I think I am mentioning something that was greatly missing from this review .. the results of the cpu on a AMD690G based motherboard. There should be a big diffrence especialy on the thermal ussage test IF a AMD690G based board was used insted of a nvidia based one.
  • duploxxx - Wednesday, June 6, 2007 - link

    fully agree on that.

    it is again a wrong comparisson. you put on of the most expensive en powerhog board to measure the performance of a low tdp cpu and put it against a mainstream competitor board.

    you can lower each powermeasurement by approx 20-25W for all the k8 series in this test.

    not even to mention the big price difference if you also use a mainstream board from AMD.

    Strange that Intel won't provide you some new e2xxx series. might already tell enough about the price/performance ratio.

    fromm older articles we alreaady know that a e4300 must be compared with the x2 4600 and a e6300 with 5000-5200. they already have the same price and in amd's favor if you add the motherboard to the combo.
  • MRM - Thursday, June 7, 2007 - link

    Maybe the tester here should look at this test

    http://techreport.com/reviews/2007q2/athlon-x2-be2...">http://techreport.com/reviews/2007q2/athlon-x2-be2...

    Here you can see the difference. Put the AMD on a nVidia board, and it will consume more power then the intel, but put the AMD on a 690g board, then the AMD will save more energy.

    So my conclusion is, if you want performance, go with Intel. But if you want a low power system ( maybe for Office, Internet or as HTPC ) then AMD is better, because it consume less power - but only if you pick the right chipset.

  • bob4432 - Tuesday, June 5, 2007 - link

    maybe i missed it, so are those power readings w/ the 8800gtx?
  • Justin Case - Tuesday, June 5, 2007 - link

    Article says "are architecturally no different <b>than</b> the Athlon 64 X2s" when it should say "are architecturally no different <b>from</b> the Athlon 64 X2s".
  • joex444 - Tuesday, June 5, 2007 - link

    As far as our language is concerned, the two are interchangable. Compare with "before surgery" and "pre-surgery".
  • Justin Case - Wednesday, June 6, 2007 - link

    No, they're not. They mean different things. "Different than" means "different from that which" (and is generaly deprecated in writing). In other words, you can use "different than" in situations like "He was a very different John than I used to know".

    Here:

    http://alt-usage-english.org/excerpts/fxdiffer.htm...
    http://englishplus.com/grammar/00000202.htm">http://englishplus.com/grammar/00000202.htm

    In the sentence in this article, it would make no sense to say "are architecturally no different from that which the Athlon 64". It's clearly a distinction between two things themselves, therefore the only correct form is "different from". Things are different _from_ each other, they are not different "than" each other.

    Your comparison with "before" makes absolutely no sense; you're comparing an adverb to a preposition. A lot of people also use "then" when they mean "than" or "it's" when they mean "its", but that does not make them "interchangeable". Not if you're literate, anyway.
  • TA152H - Tuesday, June 5, 2007 - link

    I read this article, and I'm wondering why AMD is releasing this product. It's a mediocre product by any measure. The power isn't that great, it's good, but 45 watts is still a lot, and the performance is poor to mediocre. It's inferior to the existing 3800, by and large, and it costs more. I'm just not understanding them at all.

    Why not go single core with this product, and go really low power? How many people really need dual core? If they do, don't they generally want better performance than this? It's amazing how well we did without dual processors for so long, but now they even release "energy saving" processors with dual cores. If they had half a brain, they'd halve the brains of these chips, offer them at 22 watts or whatever, and they'd have a killer CPU for the masses, and they could sell it even more cheaply. The vast majority of people don't need dual processors, or 2.8 GHz, and would love to save 22 watts, and $30 to $40 on the processor and not have it. If you're going to do something, do it right, instead of mediocre at everything. Hector has to go, he's a moron.
  • Kim Leo - Wednesday, June 6, 2007 - link

    That comment tells us quite a bit about you, first of all 45W is not a lot, and the way that AMD makes Dual core processors you probably won't magically go 22W down in consumption, more like 10W. People usually buy what is cheapest, but they are not entirely stupid, some people will understand when the salesmen tells them that 2 processors=better performance that you can feel while using the machine, that it is a better bargain.

    for anyone not playing oblivion, cell factor, or anything as requiring as those games, or editing videos proffesionally, these are top performers, hell, i'm using a A64 3400+ @ 2.4 and i have no problem playing oblivion, a little with cell factor, but i think that is because of my Geforce 7600Gt, but still. You don't need Core 2 Duo's to do anything that a Athlon(or BE 2XXX)Can't do too.

    you made a comment before this wich once again, tells us who we are dealing with..

    on the technology side you claim that P7(or the Pentium 4 design) is ahead of the K8 design, it's different, much different.. but i wouldn't say superioer in any way, the only smart thing P7 delivered was QDR bus, and even that is pretty absolete compared to HTT in the K8 design, Hyperthreading might be advanced, but it dosn't work well in any way, so if anything it is a broken advance technology(i use HT dayily in my laptop). i think that most people with some insight will know that the K8 is a wonder when it comes to technology with HTT, and IMC, and K10 will be even more wonderous looking at the power management of each core, and Memory controller, seperatly other than that it looks a lot like what intel did to the Core design, making it "bigger" with more cache and some arch changes that essentially is just making the old "bigger".

    AMD did release a 9 W singel core sempron by the way..
  • TA152H - Wednesday, June 6, 2007 - link

    OK, a few things, that say something about you. You have really poor reading comprehension, and I don't like being misrepresented.

    I never said 45 watts is a lot, I said it isn't that little. There's a big difference. When the point of a product, the only selling point really, is it's low wattage, and it isn't particularly low, it's not that great. 45 watts is just not very low. With regards to halving it, if you halve the transistors, you halve the power draw. Now, one thing is, you don't have two memory controllers, but considering everythning else, you'd come pretty close to losing half the transistors, but not completely half. So, let's say you'd lose 40% power instead of 50%. Now, you could probably run the processors at lower voltage, since you don't need BOTH processors to run at that lower voltage/speed, so, you'd still come pretty close to 50%. It would be way more than 22% or so. It's transistors that use power, and the power doesn't magically disappear into the air. If you halve them, you halve the power use if it's a carbon copy you're eliminating. Of course, as I said, it's not completely half, but it's pretty close.

    Two processors are better performance you can feel? Does one processor control a hand that massages you or something? Very, very few applications require more CPU power than one processor can deliver, and the bottlenecks in performance are more I/O like hard disks, internet access, etc... A second processor doesn't help this, but it sure uses up some power. When it speeds up surfing, text messaging, and printing then I'd say the world needs it. It doesn't.

    You completely missed my point on games. I"m not a simpleton that plays games all day, so it's not important to me, and I mentioned in a previous message that AMD processors are fine for most people, and their so cheap they are worth it. But, this particular model is confusing. It's worse than the 3800, and not very different. What's the point? They should have made something a little more interesting, instead of making a processor that's not particularly good at anything.

    Again, your reading comprehension fails you. I said the P7 was more ADVANCED than the Athlon, not superior. And, of course, it is superior is some ways, you only need to look at clock speeds for that. But, overall, it's a perfect example of a very advanced design that just didn't work well. So, again, don't put words in my mouth.

    Point to point technology is nothing new, integrated memory controllers are very old and if you don't know this you probably should read more and post less. Even in the x86 world the NX586 that was out in the mid 1990s had an integrated memory controller, so there is nothing at all advanced in the Athlon. It looked fine when Intel had the crappy and bizarre P7, and all the idiots of the world swore how wonderful the IMC was and thought Intel couldn't do it, without realizing it had severe tradeoffs too. Now the Core 2 has destroyed the Athlon and AMD, and people magically see the IMC wasn't the panacea everyone thought, at least not on the current lithographies. Intel could have done it any time they wanted to, it's nothing new or novel, but they decided it wasn't the best way to go yet. In a few years, they'll do it, finally, when it makes more sense because of smaller lithographies.

    You clearly are confused about the Core 2. It was wider, and I guess that's what you mean by bigger, but it also had memory disambiguation which is very helpful for increase scheduling. The instruction fusing and the enhanced SSE is nice too, and neither would fall under bigger. AMD is doing some interesting things with Barcelona, and the incredibly primitive load scheduling of the K7 is being upgraded to something similar to the P6 (of 1995) so that should help some, although they still trail Intel. I wouldn't call that bigger at all, actually, very little of the Barcelona is just more of the same. It's wider in terms of getting data in and out, but the processing cores are quite similar. It's more tweaked and optimized than bigger, and I'm not saying that as a negative. The primitive scheduling of the K7 is a huge handicap that even allowed the much better designed/smaller P6 to compete with it clock normalized, despite having serious disadvantages in many areas. It's still behind the Core 2, but it's narrowed the gap a lot, and should be the most important change.
  • strikeback03 - Friday, June 8, 2007 - link

    Here on my desk at work is a system with a A64 3500+, which is 2.2GHz, 512L2, and 1.5GB RAM. Down the hall in another of our labs is a system with an A64X2 4200+, which is 2.2GHz, 512L2 X2, with 4GB RAM installed though XP cannot address all of it. The dual core system is noticeably quicker in application loading and web browsing. So not something vital, but nice to have.

    Also, a number of laptop processors would seem to meet the criteria you have set (single core, TDP ~25W) though they are not cheap.
  • TA152H - Saturday, June 9, 2007 - link

    That's probably got more to do with other things than dual core. Don't forget the extra memory is used as a disk cache, and anyone that tells you surfing is CPU based is a buffoon.

    I have several of each, mainly dual processors rather than dual cores, and the differences in every day stuff isn't there. Consequently, my main machine is a single core processor because dual wouldn't make any difference for the simple stuff I do on it. Naturally, my development machine is not a single processor, but what percentage of people are programmers?

    The problem with Intel's laptop processors is that they don't use the same motherboards, so it's not that helpful. I think Aopen makes some desktop parts that will use them though, but you're stuck with an obsolete chipset if you do that. So, it's an option, but there are some tradeoffs with it beyond just the lower clock speed and cost of the processor.
  • SiliconJon - Tuesday, June 5, 2007 - link

    This is the first big disappointment I've come across here at Anandtech in the way of an article. Generally I find the articles to be very precise and technical, but I feel the comparisons are quite useless in this test setup just looking at the power consumption graphs. I really think this should have been more thorough, or at least much more precise and accurate in comparing technologica counterparts. Though it was interesting to see how far behind AMD's performance bar has fell. Where the heck are they going?
  • TA152H - Tuesday, June 5, 2007 - link

    I'm curious, what would you have considered better options to compare these processors with? Certainly not the Pentium D, since they are anything but efficient in terms of power. That leaves the Core 2, which it was compared with. Were you looking for other models in the Core 2 line or something? I'm just a bit unclear.

    Where AMD is going is pretty clear, isn't it? The Barcelona/Agena and later Fusion. HT 3.0, etc... Is it enough, I don't know. Clearly it's better than the K8 in term of memory load scheduling, but still not as good as the Core 2.

    I never liked Hector Ruiz, and always thought Jerry Sanders was an incredible CEO. He was widely criticized because the company wasn't extremely successful, although he was responsible for the success the past few years because it was his leadership that made these products possible. Every other company that fought Intel died, only AMD survived, and he deserves credit for that. Ruiz seems to have no vision, and he has to go. There is no way the K7/K8 should still be around this late in the game, and they should never have gotten so comfortable with assuming Intel would always screw up. The K7 is now almost 8 years old and outside of the minor changes made for the K8, it's still way too old and outdated. The memory scheduling is worse than the Pentium Pro of 1995 vintage. Ruiz needs to go! If this ass-clown malingers for much longer, AMD will be belly-up.
  • Regs - Tuesday, June 5, 2007 - link

    Marketing.

    There's a lot of people who think dual core is better or a lot of people who like to think dual core is "future" proof. Besides, Intel can do just the same with a single core...they all ready did.

    Bascially it's new ..hot of the presses.. and AMD want's to sell it and people like new.

    Lets face it, AMD is not going to regain the performance crown anytime soon. So they're going to sell whatever they can.
  • TA152H - Tuesday, June 5, 2007 - link

    It's unclear to me how this is new, and I'm not talking about AMD taking the performance crown.

    Your supposition that this will sell because it's new isn't altogether clear to me. It's really not even as good as the 3800 and not very different really. A single core at 20 watts would have been something new, and interesting. Or a 1.4 GHz model that takes 8 watts, or 10 or whatever it would take. I could actually get rid of my beloved Tualatins if they came out with something like that. But no, they come out with a 45 watt part that doesn't really impress anyone in any way. At least do something well, if not performance, then power.

    I agree with you on one thing though, perception is more important than reality, and dual core is perceived as broadly better, even though for many things it is useless, and almost completely unnecessary for 95% of the people. But, perception matters, so you make a good point. The P7 was born from this as well, so I guess this isn't that bad.
  • cornfedone - Tuesday, June 5, 2007 - link

    ...Anandtech jumps right to the top of the FUD spewers.

    The Anandtech statement below is good for a laugh by anyone in the PC industry with a clue:

    "Those hoping for nail biting, teeth clenching battles should apply elsewhere - the CPU war these days is a one horse race. If reports out of Taiwan are to be believed, initial performance results from AMD's Barcelona fail to impress and we've got at least a quarter before the race can even potentially get competitive. But as we've seen lately, you don't need chart topping performance to bring excitement to the game."

  • Anand Lal Shimpi - Tuesday, June 5, 2007 - link

    I suggest reading the paragraph immediately following that one:

    "By aggressively cutting prices, AMD actually made most of its product lineup below $300 competitive with equivalently priced Intel offerings. Granted that AMD won't be making a tremendous amount of money by doing this, but the end user stands to benefit, especially those with Socket-AM2 motherboards looking for faster CPUs."

    It is true that Intel does currently hold the title for fastest desktop CPUs, but with the recent price cuts AMD is competitive (if you don't overclock) below $300. That's all I was trying to say.

    Take care,
    Anand
  • TA152H - Tuesday, June 5, 2007 - link

    I agree it's a little overstated, because AMD processors are terrific for the vast majority of people, and their prices are terrific too.

    I think he meant to say (and I apologize to Anand if I'm putting words in his mouth, because I hate when people do that to me) is the PERFORMANCE DESKTOP CPU war, etc... I think most people will understand that is his meaning, and he's right. Obviously, again, most people can get by extremely well with AMD's processors, and the servers by virtue of their superior system architecture still do quite well in four socket systems vis-a-vis Intel products. But, right now, the Core 2 is generally a much better processor than the Athlon 64. In a few, rare instances it is not, and in most instances it doesn't matter. But, it still is in most.
  • SilthDraeth - Tuesday, June 5, 2007 - link

    I can't remember a single post you have made that was anything but trolling.
  • lopri - Tuesday, June 5, 2007 - link

    It's really puzzling to see all these articles published @AT today. Are you guys trying to catch up with the delayed articles? (Then again the UVD issue is fairly recent, I think?)

    First of all, we have E2140 and E2160 to compete with the AMD's new offerings. They have been available for more than a couple weeks now so I am not sure why this article is dealing with new AMD CPUs (which are not yet availble) and different class Intel CPUs. If anything, I would think these new offerings from AMD are meant to compete with E2xxx series.

    Also I would like to see NF680i along with P965/NF590 in the power consumption chart. For all purposes and intents, that'd give a better idea where things are standing. P965 and NF590 are completely different in its function/capability as well as target audience. Another consideration is that people who purchase these low-power CPU will probably look for more power efficiency when they make decisions on motherboards.

    (scratching head..)
  • Anand Lal Shimpi - Tuesday, June 5, 2007 - link

    I'm not sure what's puzzling about today's articles? Gary is in Taiwan reporting on Computex and the NDA lifted today on the X2 BE-2350, which is why you see these articles on the front page.

    The E2160/E2140 will be good competitors to the new BE chips, but we simply don't have any in-house to test, and as I mentioned above the E2160s appear to be selling for close to the same as the E4300. I'd actually say that today, the BE-2350 is more a competitor for remnant Pentium Ds in the marketplace than anything from the Core 2 lineup.

    I don't think it makes sense to go higher end in the chipset comparisons, one thing I'm working on now is comparing integrated platforms to look at power consumption (and performance) for those systems that won't be used for high end gaming.

    Take care,
    Anand
  • TA152H - Tuesday, June 5, 2007 - link

    Anand,

    I completely agree, you should be testing IGPs with these low power processors, because that's what will almost certainly be paired with it. Or, in the worst case, fanless discrete video cards. In fact, I don't think it would be completely out of place to build a whole system to be as quiet and low power as possible. You might even want to take a look at the VIA/Centaur chips. I bought one of these mules about a year and half ago, but it's just so slow even in power/performance I went back to a K6-III+. They are probably better now.

    I know you guys are allergic to fanless power supplies, but they work really well (I love them), and if you pair it with a low power system (which is only natural), they are terrific. You pair these low power processors with an IGP, or low power discrete card, stock voltage memory, and a notebook hard disk (I did this with my VIA, and it saves a ton of heat and isn't horribly slow) and you have a really low power system that is quiet. If you listen to music on your PC, or TV, or movies, this is an ideal setup. Even the alien zappers must do this sometimes, so I think it would make for an interesting review, especially during the summer when these machines can make rooms really hot.
  • lopri - Tuesday, June 5, 2007 - link

    Then why didn't you include Pentium D's in the charts?
  • Anand Lal Shimpi - Tuesday, June 5, 2007 - link

    I mentioned the Pentium 4 based processors on the front page in this quote:

    "While you can get older Pentium 4s for less than $100, you wouldn't really want to from a power and performance standpoint."

    We've always recommended the X2s over the P4s, thus the real question is how big of a gap is there between these X2s and the lowest end Conroes.

    Take care,
    Anand
  • lopri - Wednesday, June 6, 2007 - link

    You're mixing things up. I do understand where your argument is headed, but unfortunately the presentation isn't exactly relecting your intention.
  • dm - Tuesday, June 5, 2007 - link

    It overclocks well (http://fanboyreview.blogspot.com/2007/03/brag-anot...">http://fanboyreview.blogspot.com/2007/03/brag-anot...

    And the best thing is, it should be really affordable ;)
  • Spoelie - Tuesday, June 5, 2007 - link

    What was the point of increasing the latency of the L2 cache for the 65nm process if AMD isn't increasing cache size?? Now, as proven in this article, we have situations where 2ghz 90nm parts are outperforming 2.1ghz 65nm parts, and the K8 is in dire need of any performance it can extract.

    Sometimes I think AMD is doing it on purpose, they're now at a point where I'm starting to lose confidence in any sort of comeback. Underperforming r600, delayed barcelona, delayed agena, delayed hd2600xt, initial reports on all upcoming parts "disappointing", process technology 1+ years behind Intel. Way to piss away your 3 years head start with hammer AMD.

    I know this is the kind of post that gets downrated but as a long time ATi/AMD believer, the bad news just keeps piling up.
  • TA152H - Tuesday, June 5, 2007 - link

    Because, AMD will probably increase the cache size in the future. They indicated this earlier. Even now, it probably increases yields a little, and thus allows lower costs. I'm not sure where the bottleneck is on AMD processors though, but you probably remember the Coppermine and the problems it had getting over 1 GHz. Actually, Intel did release a successful Coppermine at 1.1 GHz, but it was a Celeron. This is because the problem was with the L2 cache not being able to hit the higher clock speeds. Considering this, and the fact AMD did add wait states to the cache in the event they increase the size, it is very possible this is improving yields.

    AMD never had a three year advantage over Intel. Intel mobile chips were always better in power/performance than anything AMD, just as the Pentium III was. The Athlon and Athlon 64 were not particularly good chips, it was just they were competing against a particularly bad chip (P7, of course), and when Intel ditched it, naturally things got worse for AMD. Also, The P7 was way ahead of the Athlon 64 in technology. It was a much more advanced chip, so even in that regard the Athlon 64 wasn't ahead. Granted, it worked better, but it was a much more prosaic design.
  • Calin - Wednesday, June 6, 2007 - link

    Pentium !!! might have been better in performance per power, but was worse in performance per clock (performance per power was a metric hardly used back then on desktop).
    The K7 chips (first Athlons) were faster clock-per-clock than Pentium !!! (the 100MHz FSB Pentium !!!) by at least a little. Also, the Athlon64 were better than the AthlonXP it replaced.
    True, the Pentium4 C were very good chips, even compared to Athlon64 (teh 2400MHz were very good). Too bad the 2MB cache P4 chips were even worse than the Northwoods per clock, and could not scale in frequency.
  • TA152H - Thursday, June 7, 2007 - link

    The Coppermines were better per clock than the Athlon in most benchmarks except pure x87. Read some old articles if you don't know that. Considering they were much smaller, and used much less power, and had terrible memory bandwidth, this clearly shows what a bad design the K7 was. It was mainly because of the better OoO scheduling of memory loads, although the fast L2 cache helped some (mainly because the memory bandwidth was so poor, and the L1 cache relatively small). The K7 was bigger in almost every way, and had much better memory bandwidth, yet was slower than the Coppermine clock normalized. Of course, they could reach higher clock speeds.

    I have no idea what a Pentium 4 C is, probably your misnaming the Northwood since people like to think it was good. None of the Pentium 4s were worth a damn. They were really slow for their size and power, and never compared well to Intel's mobile chips based on the Pentium III.
  • johnsonx - Tuesday, June 5, 2007 - link

    It should be noted that, aside from the TDP rating, the BE-2300 and BE-2350 are equivalent to the existing Brisbane core Athlon64 X2-3600+ and 4000+, respectively.

    One might argue that this review was unnecessarily detailed given the fact these processors are nothing more than the old ones with a lower TDP and new names.
  • Lonyo - Tuesday, June 5, 2007 - link

    quote:

    At the price points AMD is targeting with the BE-2350 and 2300, Intel doesn't really have a good competitor. While you can get older Pentium 4s for less than $100, you wouldn't really want to from a power and performance standpoint. The closest Intel has to offer is the Core 2 Duo E4300, which we've been able to find online for $113.50, thus making it the best competition we can find.

    Try looking for the new Pentium chips then. Although you don't seem to have been given any for review, they are out there, and in stock at some e-tailers worldwide. And from reports elsewhere (well, one) they seem to be decent performers, and good overclockers.
  • Anand Lal Shimpi - Tuesday, June 5, 2007 - link

    Intel won't send us any of the lower end E series chips, but from my search it doesn't look like the E2160 is actually much of a value, it's barely cheaper than the cheapest E4300 and far more difficult to find at this point. All of this should hopefully change as availability increases, but for now the E4300 is the better buy.

    Take care,
    Anand
  • Omega215D - Tuesday, June 5, 2007 - link

    Newegg has them in stock along with the lower 2140. IMO I don't see any reason to buy them since they are only $20 - 30 less but have only 1MB cache while the E4300 has 2MB. I'm also pretty sure that when you overclock both of them the E4300 will definitely come out ahead.

    While I'm here, I currently have a Socket 939 PCIe mobo with a A64 3000 in there. I was thinking of upgrading to an Athlon 64 X2 3800 ($83) along with another 1GB of HyperX RAM ($91). With the upcoming processors from both teams coming out and DDR3 should do the upgrade listed above, upgrade to a Core 2 Duo for now or just wait for the next gen stuff?
  • Calin - Wednesday, June 6, 2007 - link

    DDR3 won't come down fast in price - but for now, DDR2 is cheaper and faster than DDR.
    Depends on how much money you want to pay - for $200, a new processor and extra RAM would be the solution. If you want to pay more, and can get some money from the old configuration, maybe a cheap Core2Duo, overclocked, would be the better solution.
    I don't know what price drops are in the future, but maybe a slow, 4MB cache Core2Duo (E6320) would be the best solution - depends on available supply
  • SilthDraeth - Tuesday, June 5, 2007 - link

    You show the new BE processor using more power than the x2 5000+ at idle.

    Should it be the other way around?
  • defter - Tuesday, June 5, 2007 - link

    Yes, but because of lower voltage and not because of lower nominal clockspeed.

    Many people forget that when Cool&quiet is enabled, K8 CPUs will decrease their clockspeed to 1GHz. Thus, it doesn't matter much wherever the original speed was 3GHz or 2GHz, in both cases CPUs will be running at 1GHz while idling.
  • SilthDraeth - Tuesday, June 5, 2007 - link

    I understand that, but it was my understanding that the BE is running at a lower voltage than the 5000+. The older EE 35W is the lowest for AMD out of the three reviewed.

Log in

Don't have an account? Sign up now