Slashdot Mirror
Cyrix III Benchmarked
electricmonk writes: "Tom's Hardware has just posted their review of the Cyrix III. They benchmarked it against the older Cyrix designs, and a Celeron, and the Celeron beat the crap out of all of them. They aren't meant for desktops, however, so it really isn't a valid comparison. But it is very overclockable, and runs so cool that it can work without a fan. Quake III on an Internet appliance, anyone?"
I strongly agree with your point. This is what happens when you don't change the MS Excel plotting defaults. You should change your name to Dan (the visual-data presentation) Jedi. Then you can talk like Yoda, too.
Keeping
The grammar in the post to which you are replying is impeccable. Why didn't you understand it?
OTOH configuring an Intel based (BX or 810) is a snap, no compatibility problems, no misterious hangs, no memory rejection or disk driver malfunctions.
I just hope that the Cyrix3 is stable enough with cheap motherboards (if you suggest I buy better, pricier motherboards why bother with Cyrix?, go PentiumIII or even Duron with quality mobos, i think Celerons won't cut it anymore)
I strongly agree with your point.
Yes. My Jedi Rikert Scale Technique scores another one.
You do realize, incidentally, that one green muppet does not a stylistic monopoly create.
--Dan
does anyone know what's happening with osm? i was out of town for a few days, this is news to me. If everyone's clueless, then maybe we need a site to report on internal affairs here at slashdot... hmmm...
Hello jonnythan.
We've become quite the meta-grammar nazi, haven't we.
You repeatedly and eloquently point out errors in my grammar. I think that these are not errors in my grammar but probems with your inferiority complex. You must want the grammar nazi position real bad. I'll inform you of something Mr. jonnythan. Being grammar nazi isn't about good grammar. Being grammar nazi isn't about bad grammar. It's about mistakes. It's about an improved and grammatically pure Slashdot community.
You are wrong about one thing Mr. jonnythan. My title is not self-assigned. Rob Malda asked me to be the grammar nazi ever since Conan the Grammarian retired. He knows that I can correct grammar mistakes on Slashdot and that I am very deserving of my title. This I will continue to do.
Keeping
Yes, the early 6x86 ran hotter than Hades.. But that was 1996, and that was ancient history.. The prognosticators made fun of the p54c chips a year earlier, and they ran far cooler!! Look at the MII.. Cold as ice, stable as a promise of hell from a Catholic priest. Or the MediaGX; Cold, fast enough, and zero power.. You could pack the GX into a space half the sixe of the competition without sacrifice as well!!
.sig: Now legally binding!
Check th is out, if you doubt it.
Ummm, you've long been an AMD fan? I have a few chips to remind you of, then.
I have this AMD '486' chip that fits into a 386 socket. It makes a 386-25 into a 486-50 *snicker*
I have this K5 chip. *snicker*
Or is *long time* a relative term, and you've been running Linux a *long time* too, since the 2.0 kernel release??
Rationalize your actions however you like. I'm not going to get into some grammar-flaming war with you (that would certainly be a first, no?), but I do feel the need to comment here. I may indeed have a sever inferiority complex which I place upon you, but that doesn't have anything to do with your actual grammar errors. If moderation breeds meta-moderation, cannot grammar trolling breed meta-grammar trolling? How can you claim to be "keeping /. grammar free for 3 years" when you are doing nothing to point out mistakes? I've noticed that a good number of the times you point out "mistakes," your corrections are as flawed as the original. Besides, how can anyone trust you for pointing out grammar mistakes when you know _no_ grammar yourself? I certainly do not want you proofing my papers.
:). Don't ever trust him to watch your back in Q III either; he's one of the worst players I've ever seen (at least he's honest and good spirited about it).
:).
;)
I find little of interest on Slashdot, but it is quite an amusing place to hang around and read. Your posts are pretty funny and give me something to smile at on a pretty regular basis, so do not get me wrong. But...come on. "It's about an improved and grammatically pure Slashdot community." Where does this grammatical purity hide? I have yet to see it, and your posts are far (far, far, far) from grammatically pure.
Malda is a pretty cool guy - I talked to him pretty casually over a game of quake at the Boston Geek Pride Festival a couple of months back (the quake tournament which I should've won, but was defeated in the second to last round by a 13 year old we found in the crowd to set up our servers who cheated rather flagrantly). However, I'm not going to trust Rob with assigning us a grammar nazi
Hmm..this strayed quite a bit off-topic
You point out the mistakes other people make, and I'll point out yours. It's still a free Slashdot, no? If you really care about an "improved and gramatically pure Slashdot community," you wouldn't mind
BTW, my name is Jonathan Cousins. I live in New Orleans in the summer and spend the rest of my year at an engineering school in upstate New York. So, if you want to send the Grammar Youth after me, you know where to find me .
The big downside about considering a Duron CPU is the fact the motherboards that support Socket A tend to be really finicky when it comes to a decent power supply (you want at LEAST a 250 watt ATX 2.03-compliant unit).
I think I'll stay with a Socket 370-based motherboard until the motherboard manufacturers find a way to reduce the power consumption of Socket A-based motherboards.
Raymond in Mountain View, CA
At 0.18um going to 0.15um? No f***ing way! The main reason (aside from cooling) that 386s and 486s are used on space flights and not anything newer is that the finer geometries are too sensitive to radiation. 386s and 486s have large enough transistors and wide enough wires to have a hope of being resistant to the radiation.
The problem is, as you shrink the transistors, the amount of charge required to flip a gate goes way down. They say when you get down to these finer geometries, you get to know your electrons personally. ;-) (IIRC, at 0.07um, they figure you have around 100 electrons to flip a transistor.) The odds of a random alpha particle causing a calculation error in one of the newer chips is therefore far greater than it is in one of the older chips.
--Joe--
Program Intellivision!
This is a discussion up with which I am completely fed.
Can it be radiation hardened easily, perchance?
First off, I never say "Intel sux! M$ sux! BugBusiness sux!", but I also never get modded up. /.'ers out here are saying that Cyrix is an option they will not exercise, because it does not fit their needs.
Second, you're making a small fallacy here: If A is B, then Not-A is not necessarily Not-B. That is, just because, say, Microsoft is bad, doesn't make Amiga good. Badness is not exclusive--Microsoft and Intel may have monopolies on Software and Hardware respectively, but neither has successfully monopolized Badness yet, and I am sure neither is in a hurry to.
Third, the "Cyrix Sucks" argument raised in these forums follows directly from performance analysis, and not from an incomplete list of facts. Sure, you could use a Cyrix in your laptop--but why? It's slow! Sure it's x86 compatible--but it performs those ops slowly! Sure it's cheap--but you could get far better for barely more! Sure it's non-intel--but it's atrociously awful in every way that 'slashbots' care about! (ie. Quake 3) You are as free as the rest of us to want a choice and to use what fits your needs. However, the
holy crap! osm is one of the main reasons I visit slashdot. I was just fleshing out a front end to filter out all the +2,3,4, and 5 posts, just to read the trolling more easilly. If slashdot is really screwin with osm, that will be unforgivable. I already feel like a shmuck for recommending the site, only to have people load up the site and see this crap. i haven't been posting much lately, but that don't mean i ain't hittin banner views... I will "boycott the dot"(c) if any of this turns out to be what I think it is. If they sent a cease and desist to osm, it will be (civil disobedience) war!
"You must want the grammar nazi position real bad"
any one who REALLY knows Engrish, knows that -real- should be -REALLY- in the above sentence. You know, for a 'grammar nazi' you are REALLY an IDIOT. Your little bullshit story about being 'selected' is just so much daydreaming in calculus class, isn't it kid.....
Some of your grammatical mistakes would make your teachers blanch and faint, "Grammar Nazi."
it's alreaedy been done. trolldot
Does anybody else second my sentiment that Seagate's drives have been the noisiest going all the way back to the ST-225?
Has anyone done a symphony for harddisks yet? ST-225's for their rattling noise, kalok 40mb's MFM for that chirpy sound, a few whining full height micropolis SCSI's... oh wait... maybe i should just tape this right here!
It worked more or less ok for about 6 months, but after that I started crashing left and right. By this time I had another primary box, so I threw a used Intel chip onto the board to see how it reacted. Up time increased by a factor of ten!
This should be a great thing for those of us who are driven absolutely crazy by fan noise from computers.
There was an excellent Ask Slashdot on this very subject in November and another one (can't find the link) recently.
MIPS is a notoriously unreliable measure, because different architectures do a different amount of work with a given number of instructions.
New metrics that would probably be useful:
These benchmarks are straightforward to compute, given an appropriate testing rig.
For those of you who didn't know, the Cyrix III was originally supposed to be based on the Joshua core. It was supposed to be a highly integrated chip. It had a two way superscaler core, and was very RISC-like. It was supposed to debut at 600+MHz at a cost of only about $60. Additionally, it was to have integrated Voodoo-2 level graphics and a memory bus capable of transferring 3.2GB/sec using multiple channels of RDRAM. For more info, you can look to issues of MaximumPC where Tom Halfhill wrote an article about he proc.
A deep unwavering belief is a sure sign you're missing something...
The Cyrix III's biggest competitors may turn out to be National Semiconductor's Geode, Transmeta's Crusoe, Rise's newest products and even Intel's impressive StrongArm. ;))
Have there been buyouts I haven't heard about? I'm pretty sure that it's Cambridge(UK)-based Acorn that manufacture the StrongARM chips, not Intel. Of course, IANAL. Not that it would matter if I was, but that acronym is fast becoming a pre-requisite for a slashdot post.
Blaming GW Bush for the Iraq war is like blaming Ronald McDonald for the poor quality of food.
I am only 16 years old..... even 4 years is a long time for me. So, go to hell you obsessive compulsive social reject!
Buying a Dell computer is equivalent to dropping the soap in a prison shower.
I remember when Cyrix was THE competition for Intel. AMD were the little suckers with slow chips, and Cyrix were the runners up that got swamped every time.
Now AMD is big, Intel is big too... and Cyrix is still lagging. But they can be good, too.
With a sufficient amoutn of funding, and a good market niche such as Internet appliances that will require low power, Cyrix might find a good fight. And they might be good competition for Transmeta in this market.
Time for a bit of meta-meta grammar correction :)
.'s? What sort of punctuation mark is that? How come any post criticising anyone's grammar is always full of mistakes? Yeah, I know a few will creep into this one. Sue me.
really bad? Not correct either! Better:
really badly
Of course, which ever way you look at it, it's just a phrase; I suppose the only way you could make it acceptable is to drop the really out of the sentence: "You must want the grammar Nazi position badly". (Note that Nazi starts with a capital 'N').
isn't it kid.... 5
quantum... bigfoot... aaaaaaugh..
i have a story about trying to build a usenet server out of those (hey, 6GB was massive when they came out) on an IDE raid on linux 2.3.. wanna guess how much it SUCKED?
yes, exactly that much.
i browse at -1 because they're funnier than you are.
Why is this flamebait? What flame was it baiting? The anti-fucker contingent?
...is fabbed by Intel, according to designs by DEC/Compaq, which is involved in a cross-licensing agreement with ARM Ltd, which was spun off from Acorn Ltd early in the 1990s. Acorn Ltd is now know as Element 14.
:)
Hope that clears things up.
--
Peter
Cyrix 3's? Now I guess we know what the new Packard Bell's will be built around. Add in a Quantum Bigfoot hard drive, and I think you could officially refer to them as Diablo, Mephisto, and Baal...
The article here states explicitly that the Celeron pretty much "smoked" the Cyrix--and, probably not coincidentally, so does the page the benchmarks are posted on. Looking at the actual benchmarks, on the other hand, doesn't exactly tell the same story.
In fact, while the Celeron humiliated the Cyrix in graphic-intensive trials, the Cyrix really held its own or surpassed the Celeron's performance in the majority of those operations which did not involve a lot of pixel-crunching.
So, despite what the text of this Slashdot article says, the Cyrix may be a very useful tool, even if it won't make your frags look cooler.
Do Slashdot authors actually look at the pages they report??
"Beware he who would deny you access to information, for in his heart he deems himself your master."
With the right motherboards, the Duron will be a real winner. Maybe stick a HighPoint chip on there, to circumvent Via's and AMD's disk transfer rates which are in the crapper...and give us some overclocking options...and you've got a great opportunity for overclocking heaven if you stick an Alpha on it!
Hopefully Soyo will make a decent Duron board - the 6BA+ IV, their flagship BX model, which my 500E is on, is the best board I've ever used. It's incredibly stable even running 1.5 times faster than normal (image here), and if they make a Duron board I can't wait to see how far people take these things.
--------
Oscarfish.com: tropical fish with attitude. Way t
It will require a lot of marketing to build up Cyrixs' reputation though, and i would doubt that most consumers would even consider it as a serious brand.
Whoa. You shouldn't have written that last paragraph. It hurt my eyes!
Playing Quake 3 with fewer then 40 frames per second isn't good. I guess I'll have to stick to Intel Celery for that.
Why would Cyrix leave L2 cache off of this processor? They aren't worried about competing with one of their better processors, because they don't have any better processors. L2 cache in a cheap processor allows for things like cheap SMP computers (L2 isn't neccessary, but helps considerably). With performance ratings like what the Cyrix 3 got, maybe they should have called it the Winchip 3!
Keeping
What do you mean? Wasn't that demonstrated on the transmeta chip during the debut...I remember Linus getting smacked down on there...
I, for one, dont care about brand loyalty. I could care less about what processor I buy. I have been a long time AMD fan, only because they have the best bang for the buck. I am about to buy a 700MHz Athlon in a couple of days, only because it is half the price of equivelant Intel cpus and just as fast or faster. The day Cyrix produces a cheap processor that is just as fast as any Intel or AMD cpu will be the day that I buy one.
You can take your little "The whole lot of you" quote and change it to "'The whole lot of you' - 'Kwikymart'"
Buying a Dell computer is equivalent to dropping the soap in a prison shower.
I can, in fact, imagine a beowulf cluster of 'these muthas' as you so delicately put it. It is my firm belief that a beowulf cluster of sevearl million Cyrix 3s would acheive true intelligence, as eash node would have roughly the processing power of a human neuron. (If you're looking for a science fair project, how about this? Cheap, too!)
Early Cyrix chips ran extremely hot. Without a very good quality heat-sink they would burn up, I know, I've had to deal with badly made Cyrix systems at work. Right now two of my computers are rejects that were permanently fried after 2 years of high-temp operation.
Though I am somewhat seriously getting a Cyrix III just so I can use fewer fans. I've had enough of high-performence. :)
I was having a conversation with a friend about an hour ago about AMD and remarked that the Athlon has been very good for them in more ways than one. The success of this product will definitely attract more engineering talent from colleges and even other companies like IDT/Centaur and Intel. What this basically means that the K8 will be better than the K7, and so on. You can identify this effect already just by the improvements made to the K75 (Athlon Classic 0.18mu) core for the Thunderbird/Duron product lines.
I think this definitely involves VIA and Centaur. I wonder what happened to those engineers who worked so hard on the Joshua core only to have it scrapped in favor of this Centaur core. Sure, it's darn skippy that it'll run on 10W and produce next to no heat, but (as another poster remarked) this is a desktop, not a handheld CPU. If I were VIA (and I'm obviously not
VIA has done some nice work in the past, but IMHO they have their heads up their respective rears. The memory performance of the Apollo Pro 133(A) and KX133 chipsets has been downright awful up until very recently, and even now is still slightly less than the performance of a BX-133 (and even BX-100) platform. What good is forming a new standard (PC-133 and soon PC-1600 and PC-2100) if your chipsets *barely* support it? Not to mention crummy ATA66 disk performance and this bogus 'incompatibility' with the SocketA platform. Pfaugh. If any corporation stands the risk of becoming "the next Intel" it's VIA, not AMD. Truth be known, the only reason VIA is alive today is NOT because of AMD, but because of the Intel i820+Rambus fiasco. That's my $0.02 at least.
At any rate, back to my original point. If you were an computer/electrical engineering student with a career path leading straight to Intel, would the recent 'outbreak' of Transmeta, AMD, and VIA change your mind at all? I think so. I also think that many of the engineers at Intel are considering a lateral career movement. It's just a suspicion though.
Food for thought.
Alakaboo
"Yoda you talk like." What is that? Yoda uses correct grammar, save for some dangling participles. 'Yoda you talk like' is a travesty of grammar, if I can use the term so loosely.
Perhaps "Talk like Yoda you do" is the sentence you were looking for?
I agree totally. If this is truly what has happened, and this would suggest something horrible has happened to osm, we should not only actively boycott slashdot for this behaviour, but email taco every day that we do not go and explain why, report to other news sites, post on other news sites in our support for osm daily.
Obviously now that /. is coprporate this will be the only way to get them to do right. It is really interesting how hypocritical /. can be. They advocate open sourcing everything from toasters to nuclear codes but /code has always been woefully behind and released "when [taco] feel[s] like it." (direct quote) They talk about allowing free speech, anonymity, and privacy, but make /. all but unviewable without an account. Then they pull this shit. It was bad enugh when they wee disabling accounts for "trolling" and indeed IP addresses. It is the reason some of the newer trolls (Karma pimp? where art thou?) have disappeared.
It's not like they really encourage positive participation anyway. Moderation will automatically make you leak karma like there is no tomorrow, so the only people who will do it have vested interests. Posts critical of /. practices (like these) will be modded down. YOu can't moderate down signal 11 or you will be bitchslapped by Taco. You also cannot annoy Taco in any way or he will bitchslap you too.
At one point there was supposedly a bug in slash that allowed posts to end up at -2. Probabaly an urban legend, but interesting if it was true. Given the way posts are moderated, it would not be hard to (re)impliment silently because anyone who dared point it out would be moderated down to -2. Thankfully this has not happened yet.
Maybe we should support and send articles to the troll-friendly slashdot2.org? Perhaps that page could become the place to report bullshit on /. since /. no longer reports on it anymore (they used to report when they were hacked, DDOS'ed, down, or anything, now they are down in some fashion almost daily with no explanations, and all kinds of nefariousness goes unexplained, like the undocumented bitchslap system).
Has anyone done a symphony for harddisks yet?
Several people have tried, but the hard drives have a nasty tendancy to head crash during the second movement. Head crashes are supposed to be saved for the big finally, damn it!
Those are the most disgusting charts I've ever seen:
.5 FPS? Are you kidding? (No, I didn't count exactly fifty pixels. So sue me.)
Tom can say what he will about RDRAM, and nVidia, and 3Dfx, and whatnot. I'll be amused, but I'm not going to get pissed.
These charts piss me off.
Half a frame per second lost from AGP Fast Writes in one game does not a half-chart spanning differential make.
Graphing two values against eachother is meaningless if the scale is not consistent from graph to graph, you just end up with "more" vs. "less" being visually amplified, without "perceptably equal" even being an option.
Fifty Pixels Of Hype over
Yours Truly,
Dan Kaminsky
DoxPara Research
http://www.doxpara.com
I want to see more charts like this. Nice.
Yours Truly,
Dan Kaminsky
DoxPara Research
http://www.doxpara.com
That cyrix is not even in the desktop chip market anymore and is competing with transmeta, for internet appliances?
"Alcohol, cause of, and solution to, all of life's problems" -Homer Simpson
We need a new performance metric to sort out this mess. How about the number of MIPS per WATT?
At first I thought to myself "hasn't this guy heard of AMD?", but then you mention it at the end of your post.
so um, what gives? Just because some of us dislike Intel doesn't mean we have to love Cyrix.
well, that was kind of a bitch slap. Metrics are hardly ever relevant when taken out of context like that, though.
kick some CAD
Get a cyrix-III a Maxtor Hard drive and Win95 (OSR1). Add in 32Meg of RAM and make sure you bought the box from Packard Bell.
Your box gets it's very own darwin award!
FluX
After 16 years, MTV has finally completed its deevolution into the shiny things network
"It is seldom that liberty of any kind is lost all at once." -David Hume
Anyway, regarding our Tomas' (I prefer my spelling ubercultural like, entiendes?) "alleged" stint as an intel lackey - the proof is in the volkswagon. Schiessestrasse and Main. The keys are in the exhaust. Do pinch your nostrils before unlocking the trunk. I tried to raise a cardiac spasm but there was too much silicon in the way. Live lying down, die lying down, you know what I mean?
Finally, dont be a fool, man! Serial #s are an antitheft device; Cyrix has NOTHING to worry about.
P.S. Tomas should stick with the Dobermen. One, hooker, two hookers - it all adds up. Pretty soon we're talking an entire bordello and then things have gotten entirely too far out of hand to. That kind of lawyer action is gonna cost a shitload more banner ads than even Tomas can raise.
Hey, they're perfect for storing MP3's... and at least they wont fail on me when the temperature in the room gets over 30C, unlike the seagate 8.4GB ide in here... whoever had the brilliant ide to encase a harddisk in rubber must either have been living at the north pole or be utterly stupid...
Excellent. This should be a great thing for those of us who are driven absolutely crazy by fan noise from computers.
My computer is so fscking loud I sometimes almost want to pull out my Beretta and empty my clip on it.
Are you adequate?
VIA did a good job with price and power consumption, but what, in terms of platform, is this useful for?
If I've got a socket 370 MoBo, I'm not going to put a Cyrix 3 in it.
I think if VIA wants to make a chip for low-end use, they should provide a low-end platform for it. Like an equally cheap MoBo, built in components that might run off 12v DC. Another with a TV tuner on MoBo. If they want to market these, I think they'll have to *make* a market for them.