Slashdot Mirror
Ask Slashdot: Should CPU, GPU Name-Numbering Indicate Real World Performance?
dryriver writes: Anyone who has built a PC in recent years knows how confusing the letters and numbers that trail modern CPU and GPU names can be because they do not necessarily tell you how fast one electronic part is compared to another electronic part. A Zoomdaahl Core C-5 7780 is not necessarily faster than a Boomberg ElectronRipper V-6 6220 -- the number at the end, unlike a GFLOPS or TFLOPS number for example, tells you very little about the real-world performance of the part. It is not easy to create one unified, standardized performance benchmark that could change this. One part may be great for 3D gaming, a competing part may smoke the first part in a database server application, and a third part may compress 4K HEVC video 11% faster. So creating something like, say, a Standardized Real-World Application Performance Score (SRWAPS) and putting that score next to the part name, letters, or series number will probably never happen. A lot of competing companies would have to agree to a particular type of benchmark, make sure all benchmarking is done fairly and accurately, and so on and so forth.
But how are the average consumers just trying to buy the right home laptop or gaming PC for their kids supposed to cope with the "letters and numbers salad" that follows CPU, GPU and other computer part names? If you are computer literate, you can dive right into the different performance benchmarks for a certain part on a typical tech site that benchmarks parts. But what if you are "Computer Buyer Joe" or "Jane Average" and you just want to glean quickly which two products -- two budget priced laptops listed on Amazon.com for example -- have the better performance overall? Is there no way to create some kind of rough numeric indicator of real-world performance and put it into a product's specs for quick comparison?
But how are the average consumers just trying to buy the right home laptop or gaming PC for their kids supposed to cope with the "letters and numbers salad" that follows CPU, GPU and other computer part names? If you are computer literate, you can dive right into the different performance benchmarks for a certain part on a typical tech site that benchmarks parts. But what if you are "Computer Buyer Joe" or "Jane Average" and you just want to glean quickly which two products -- two budget priced laptops listed on Amazon.com for example -- have the better performance overall? Is there no way to create some kind of rough numeric indicator of real-world performance and put it into a product's specs for quick comparison?
As soon as someone gives me a definitive definition of what "real world performance" for a CPU/GPU is that doesn't change over time/software-version/user-care-ometer is, I might agree that it's feasible to use it to name models.
You know, like Consumer Reports? Maybe? Where we could list this stuff?
Passmark. You're welcome. https://www.passmark.com/
'Real world performance' according to who or what, precisely?
Clock speed doesn't tell you the whole story and to the vast majority of people (read as: non-technical people) it wouldn't mean anything to them anyway, other than maybe one number is bigger than another number.
Same goes for so-called 'benchmark' test suites, which I think can be argued as being biased in one way or another (or a processor gaming the system to make it appear it's faster on such-and-such benchmark test).
I think that for the people such information matters to, they're going to already know what's what without anyone spelling it out for them.
It isn't possible. Microsoft tried this with the "Windows Experience Index". It failed. dryriver asks the best questions though.
Why make it so complicated?
What exact 'performance' figure does dryriver suggest?
Raw GIPS/TFLOPS? pretty much meaningless and very easy to get an achievable peak number.
SPECINT/SPECFP? with what OS, compiler, flags, version, etc?
Anyone who knows much about cpu/gpu performance knows why this is a very very very silly 'suggestion'. It would be not more meaningful than the numbers they assign now.
The complain should be with the manufacturers - please come up with more sensible naming practices, but in the end, thats their decision.
Marketing is what is being questioned here - nothing technical - and good luck with that..
Market separation/obfuscation through complex naming schemes is seen as a feature by manufacturers, not a problem..
You've wasted everyone's time. You've used electricity to create, store, transmit, and receive this stupidity. You should probably feed yourself to the angry midgets.
I suspect Intel went to the i3/5/7 numbering because they could not continue to raise clock speeds. The new numbering obfuscates performance. For example, I'm running an i3 desktop that while 2 core, each core is faster than many i5 single cores. That means I get great performance out of a single thread at a much lower price. It's just not as good at handling numerous simultaneous processes.
Yes, and your children should be named according to their IQ.
Is there no way to create some kind of rough numeric indicator of real-world performance and put it into a product's specs for quick comparison?
If it was manufactured this or last year, it's probably better than Joe or Jane's 5-year-old laptop which was more than likely working just fine for them (modulo bloatware and registry cruft) until it broke. That's good enough, right?
Cars are also complex, they don't have simple-to-understand names and variants and require you to document yourself and investigate for large amounts of time before committing to a purchase.
Don't try to dumb down complex machinery. It will never work.
...gis sdrawkcab (usually not responding to ACs; don't bother posting as AC)
Since we're living in an ideal capitalist society, the number that indicates relative performance should be preceded by a dollar sign.
Characteristics of a CPU: fab model, fab date, fab country, transistor size in nm, frequency in GHz, num of threads, num of cores, sizes of L3,L2,L1 caches, is 64-bit?, IPC, ...
Oh yeah, I'm currently at 5808 bogomips on my I7-7820HQ.
https://en.wikipedia.org/wiki/...
I use passmark as well, but some things in the newer CPUs do not relate to the raw processing performance. For example, my tablet has a 4 core atom processor that's roughly the same passmark as a Core2Duo system. But it can decode and encode H264 video with little to no cpu utilization. Because they added that feature to the chip. So now I can watch 1080p video on a system that could not consistently decode it on the processor.
summarized; a small group of greed fear ego based inbred psychotic wmd on credit cabal sponsored throwback last gasper style genociders.. cease fire stand down.. thanks again..
Throw that Bates 6000 away, it's a piece of shit
Seeing as it took Intel so long to go from i3 to i5 to i7 processors to only now releasing i9s they have a long way to go to get back to the glory version number days of i386.
In all seriousness though, I've kind of given up on making sense of the processor/GPU models and just paste it in Google to see the specs and compare that with another one I am already familiar with.
yes. there is no way unless you want to pay. "Is there no way to create some kind of rough numeric indicator of real-world performance and put it into a product's specs for quick comparison?"
1. Like EPA fuel mileage tests, manufacturers will find ways to rig their chips so they benchmark better.
2. Unlike the EPA/feds, there's nobody to punish corporations when they cheat.
It's really not hard to do a little research to see how CPUs compare. Yes, it's a PIA if you're buying spur-of-the-moment and comparing laptops at the Big Box Store. But you need to do research. Hyperthreading and multiprocs will speed up some apps and do very little for others, some standardized benchmark number printed in the specs won't really tell consumers anything very useful. Too many variables and dependencies.
Never let a lack of data get in the way of a good rant.
Just convince them that the HDD option isn't worth the $100 in savings.
your thin skin doesn't make me a troll
In a perfect world, yes. However...
All you need to know, really.
Of course, I think these days, they have to be written in scientific notation.
The bigger the name/number, the more profit they expect to make per unit.
I'll try a novel analogy instead of the typical car thing. Imagine these chips (CPUs, GPUs, etc.) as shoes. Yes, shoes. Now there are obviously shoes of all kinds of sizes and types, and no one shoe of a certain size/type can be said to fit a particular person's requirements. Too big, too small. Great (9) for the red carpet runway, not so much (2) the tarmac kind. Perfect (10) for the alpine, chafing and sweaty (1) on the beach.
User A does spreadsheets all day, B does FPS games, C does CAD, D AI research, etc.. Some require multi-threaded performance; some, single-threaded, etc. etc.. What might seem like a good performance for one use is weak for another. It's just not possible to come up with a workable single axis performance metric when performance is determined by multiple variables, each having their own weight depending upon the user.
If you want to shop for kit that best fits your needs, you first need to come up with an understanding of the importance of each of the variables then go comparison shop the various benchmarks out there. As with most nearly everything it's best to just ignore the marketing speech and go do your own research.
Two of my imaginary friends reproduced once
I got bitten by Intel Obfuscation Syndrome when I bought a Core 2 Quad Q8200, not realizing that it was the only one of the Core 2 Quads to not have virtualization. Yeah, I should have looked before I leaped. In the end, it was a bad buy all around, as the DG43NB motherboard I bought to go with it also ended up crapping out in a surprisingly short time, but lasting long enough to be out of warranty. Needless to say, all of my later builds have been AMD (with various makes of motherboards).
Oh, no! You have walked into the slavering fangs of a lurking grue!
Somehow, game companies manage to figure out which CPUs/GPU are required and preferred for each of their games. Of course it's almost impossible to tell if my current hardware meets those specs because the numbering is completely out of order. i3, i5, i7, sure the i7 is somehow better, but how much better? Will my top end i5 beat the medium tear i7 that they ask for? It's maddening. At this point I only buy Nvidia GPU systems because I've sweated blood learning their numbering system and I don't want to figure out ATI's system and get them confused. This crap makes Windows version counting look simple.
It's the marketroids. They are laughing their ass off.
Other aspects beyond CPU/GPU specs are more important. This includes amount of RAM, size of disk, size of screen, battery life, and others that I can't name off the top of my head.
This is why you don't go Amazon and blindly buy some laptop hoping it is the best one for your needs. You talk to people, do some research and yes, read some reviews to get an idea what your buying. If that is too much for you, then your probably going to be happy with whatever you get, since you really don't care enough to put in the initiative.
Sadly, whether your buying a toothbrush, a house or a gigantic black dildo it falls on the consumer to weed out the garbage and determine the best product for their needs/budget. Computers are no different, as long as they are selling you can be certain the companies producing them won't waste effort making sure the uninitiated understand what they are buying. In fact a well informed consumer is far harder to sell with flashy advertising and misdirection than "Joe Consumer". If everyone really did their research all those gen 2 i7 laptops sitting on Walmart's shelves would rot as new products where released, rather than getting bought by someone getting a "deal". Because you know; i7, that means it's the best right?
As "Computer Buyer Joe", I have found that the best approach is to get my computer nerd nephew to hook me up with the good shit. I tell him how much I can spend and which games I want to play and he does the rest. Then, I throw him $50, which he immediately spends on oxycontin or rap records or whatever it is that kids spend money on these days.
You are welcome on my lawn.
They already do this. Always look for the standardised number following the dollar sign.
I can dive right into the numbers because I've been immersed in tech for 25 years. Put in the time to learn it or lean on someone who already has and compensate them fairly. If it's a friend or family member, do them a favor; if not, then buy the parts from whoever you talk to, or compensate them monetarily. You can't trivialize this...you can't boil it down to some simple number to describe all types of components.
I don't think the name needs to denote real performance numbers. However, it ABSOLUTELY SHOULD denote different products which HAVE PERFORMANCE DIFFERENCES! Case and point, AMD first released a RX560 which was benchmarked and reviewed by all the media/press which used 1024 Stream Processors and 16 compute units. A few weeks/months later, AMD quietly released a new version, still calling it the RX560 (with no other indication of a change and no announcement of a change), and instead having 896 Stream Processors and 14 compute units. Nothing indicates the difference between the two versions, but one version is absolutely less powerful than the other. Nvidia is now essentially doing a similar thing with the MX150, releasing a new version that uses less power, and runs with a 36% slower base clock speed, 32% slower boost clock speed, and 17% slower memory clock speed. Benchmarks are showing it having about a 20-25% slower performance as a result. Yet, it is still called MX150, and will be advertised as such when people go and look at the part.
In both these cases, they should be clearly renamed a different product, denoting the actual product, which you can then get clear idea of the performance of the product based on reviews and benchmarks.
But how are the average consumers just trying to buy the right home laptop or gaming PC for their kids supposed to cope
They don't need to. The average user will have their needs met by any computer built in the past 10 years.
if you want high-end or specialised stuff, just let the price guide you. The more expensive (so long as you don't get suckered into paying a brand premium) a generic computer is, the better it will perform.
Most people buy to a budget, anyway - not to a specification. That is why the first question a sales-droid will ask you is "how much do you have to spend?".
politicians are like babies' nappies: they should both be changed regularly and for the same reasons
The model names are mostly standardized now, even across different manufacturers.
Intel has, for example Core i7-4790K, and Core i7-8700K. AMD would have Ryzen 2950x, and nVidia would have GTX 1070 Ti. There is a similar pattern in all of them.
Intel (desktop) chips read like 4-7-90-K, 4th generation, i7, last iteration (highest performance variant), unlocked (non-K versions are not enabled for overclocking). Then 8-7-00-K would be 8th generation, i7, first iteration.
AMD copied this to an extend. 2-9-50X would ve second generation Ryzen, i9 counterpart, mid-level, but -X suffix seem to mean slightly improved performance (all AMD chips were unlocked for overclocking).
nVidia is similar 970 would be 9th generation GTX, second highest level (geared towards gamers with mid-to-large budgets), while 10-80 Ti would be 10th generation GTX, highest level (geared towards people with serious money), and updated (Ti) edition.
In general, generation increases add significant power reduction, allowing less running cost, and higher performance for the same price. In fact a future i3 might be better than a previous i5.
(I'm skipping Pentium/Celeron which are lower binned silicones of the same design, and Atoms, and of course Xeon server and workstation chips).
Looking at Wikipedia for the CPU/GPU generation gives sufficient detail for differences between offerings. If I'm planning to purchase a CPU to use for many years, I would benefit spending some time understanding those differences.
Instead of making buying decisions from info on websites (Amazon, Newegg,etc.) trying to sell you the product, why not try some other sources like hardware review sites?
Bonus Tip: The two budget priced laptops listed on Amazon.com? Performance sucks on both.
I'm so sick of performance per watt. I can figure that junk out by looking at how fast a chip actually is and it's power dissipation. What's hard is actually figuring out the raw performance of each given card. Especially since for quite a while raw performance numbers went down..
Great!
Now, I happen to use my CPU for high load database queries, and my GPU is used for offline rendering using a custom OpenGL graphics chain.
I am SO pleased that Passmark will accurately determine the relative performance of all CPU/GPUs for me...
Translation: Benchmarks are of limited use as they only represent limited workloads.
There IS no one figure that ranks performance of all CPU/GPUs for all workloads, pretty much by definition.
Aside from the fact that you can't clearly define any simple set of tasks as being indicative of "real world performance", you also can't dictate to manufacturers what they call their products. As soon as you come up with a suite of tests that is your "real world" benchmark, then you can guarantee that manufacturers will optimise their designs specifically for the suite of tests you're running and game the benchmarks.
Re: the numbers, this would be like telling Audi that they can't sell a car called an RS3 that is faster than an A4 because the number at the end is smaller. But what defines better for "real world" use in an automobile? Some people would say that a bigger car is more useful because you can fit more stuff in it. In this case, the A4 would be better than the RS3. Some people say a smaller car is better as it's more manoeuvrable and easier to park. Some would say a faster car is better. Some would say a cheaper car is better. Some would say a more fuel efficient car is better. Some would say a larger engine is better. Who is right and who is wrong?
What does the number at the end mean? It's simply a model identifier, a family name. Intel follow a fairly strict naming convention for the model names, they're not simply plucked from thin air (well, in part they may be, but that's a minor part of the name).
https://www.intel.com.au/conte...
https://www.intel.com/content/...
Specialist Mac support for creative pros, Melbourne
Do you really think you can boil a performance metric down into a single number? It's a multidimensional problem.
“Common sense is not so common.” — Voltaire
just some approximation.
I've decided to stop wasting my time responding to AC trolls/sockpuppets... so if you want a response from me... login.
nVidia is similar 970 would be 9th generation GTX, second highest level (geared towards gamers with mid-to-large budgets), while 10-80 Ti would be 10th generation GTX, highest level (geared towards people with serious money), and updated (Ti) edition.
Bzzt! GeForce 9 series (9xxx) is the 9th generation, and is rather old (10 years) and includes 9600 GTX, 9800 GTX. ... GeForce 900 (18th), GeForce 10 (19th)
GTX 970 is actually 18th generation: GeForce 9 (9th), GeForce 100 (10th), GeForce 200 (11th), GeForce 300 (12th),
The average consumer will look at three things; The CPU speed, RAM size and hard drive size. You can't assign a single spec to computer components because how they interact with each other matters. For example the motherboard's bus speed can have a huge effect on performance, but only if you have ram that is fast enough to use it and a CPU that can keep up.
Rule of thumb: Build the PC yourself. I start with price and review score. If it's cheap, there's a reason. Get a good motherboard then research out what CPU and RAM you will want that will fit it. You probably won't replace or upgrade these three components so do it right the first time. Run the OS off a SSD. If you need more space then get a standard spindle drive as a second drive. Then get a Video card that is far from the latest greatest. You're better off dropping in a new one-year-old card every year then to buy a top end one every three years. Don't worry, it'll play the latest games at full spec just fine.
I've been following this recipe for about 20 years and usually spend $700-$800 whenever I do a major upgrade. The motherboard usually lasts about 5-7 years before I'll upgrade. When I buy a game that it can't play really well, buy a new video card. When I run out of hard drive space, I add a new drive or replace an existing one.
It's crazy to me that my desktop at home feels much faster then my work PC even though the specs of my work PC are much higher. There simply are just so many factors that make up performance then it can't be accurately tracked with any one metric.
https://en.wikipedia.org/wiki/...
"When a measure becomes a target, it ceases to be a good measure."
"All metrics of scientific evaluation are bound to be abused. Goodhart's law (named after the British economist who may have been the first to announce it) states that when a feature of the economy is picked as an indicator of the economy, then it inexorably ceases to function as that indicator because people start to game it.[6]"
Why should GPU's be treated any different than any other product. Car names have no bearing on performance, drug names CERTAINLY have no relation to what they are or do to you or for you. It is strictly a marketing scam to try and grab the attention of potential buyers. Consider Ultra Mega Platinum Extreme vitamins. I could see a case where they should be required to provide some actual real world indication of mflops or tflops or general productivity but that would require some standardized measure performance, and as long as the architecture is different and the drivers proprietary they could arrive at the same output using different processes.
errr....umm...*whooosh* *whoosh* Is this thing on ?
This is a problem device manufacturers should be willing to fix. If consumers cannot decide which product to by on performance, the only way to decide is price. Such a market is doomed to crush manufacturer's margins.
But fortunately, most consumers are OEM, which may still have the ability to understand parts performances
"I made some uninformed decisions so now I don't buy that brand anymore" is a ridiculous viewpoint.
I don't think trying to make CPU/GPU numbers comparable between vendors is a good idea - whatever standard is used WILL be abused and exploited, to the detriment of actual performance if need be.
But within each vendor, there should be general ways to tell performance based on a model number and a simple, consistent numbering scheme.
* Some number needs to indicate relative performance. A higher number here should indicate higher performance in every reasonable usage. These do not need to be on an absolute scale - eg. if a CPU700 is twice as fast as a CPU500, a CPU600 could be anything from 110% to 190% the speed of the CPU500.
* Some number should indicate generation or featureset. If you added Feature X in Generation 2, there should be a 2 somewhere in the model number, and anything with a 2 or higher in that spot should have Feature X. Socket compatibility might also be good to put here - one ought to be able to say "this motherboard will work with any [vendor] CPU starting with [digit]", which is sadly not currently the case.
* Certain features ought to be knowable given only the model number. For CPUs, core count is important - I really hate how Intel will label dual-core mobile chips as "i7" or "i5", markers that absolutely indicate core/thread count on desktop. And it seems AMD is copying them now. On the GPU side, I would go with either CU count, as AMD is doing with Vega, or memory bus width.
The Cpu manufacturers will just find some way to game any kind of Benchmark. For such a Benchmark to be usable it will have to be relatively unchanged for some time. They'll just focus on how to game whatever the benchmark is testing while letting the CPU stagnate in other areas.
The only way to know for sure is to compare reviews of the chip on a similar workload you're planning to run on it with several competitive benchmarks and see who wins out on average
This entire post is based on a faulty premise. Nobody selecting components for a PC should be so dense that he can't identify that one number is greater than another. Anybody asking this question should just buy a prebuilt system or, better yet, a Mac so that he doesn't waste the time of competent people with this bullshit.
The warning tag inside the airbag clearly said it was filled with razor blades. Not their fault your face looks like a tic tac toe board.
What's the difference. Aren't they both benchmarking apps? I used passmark scores to decide on a CPU during my last round of upgrades and it seemed pretty in line with what I experienced when I owned the CPU.
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
but laptop performance varies wildly based on the efficiency of the cooling solution. On my work laptop I went from a computer with a slower CPU to a faster one spec wise and took a pretty big performance hit even in stuff like Excel/Word.
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
Windows Experience Index was one of the most sensibly designed consumer benchmarks I've ever seen. By running multiple subsystem tests and using only the worst single-test score as the overall system score, it rewards well-balanced system design and penalizes cheaping out on specific components.
It goes back well before this.
Once upon a time processors were labelled with their generation and processor speed (eg 80486 50Mhz). Then around the time of the first Pentium the marketeers started to worm their way in there and explicitly tried to confuse matters. It went downhill from there.
For a while AMD named their processors as suggested, relative to the Intel speed. But in the end the marketeers wormed their way in there too.
The issue is that if Intel labelled their processor with it's actual performance then you wouldn't pay $200 extra for 5% extra performance - this way they can hide how small are the real world gains.
If so, then put a 2-digit number after your name to indicate your real IQ. Duh.
Generally speaking can't choose their GPU/CPU combination at least not fully. They are buying a complete machine. They have their price point, they have base needs that will vary slightly like must have hdmi, or must have a serial port for that damned thing they can't give up but for the most part: "I'm willing to pay $700 what can I get?" they then look at most look at about 3 options from 3 different vendors say Asus, HP and Dell. They then end up with whatever ram, hdd, CPU, GPU combination that is "best" (however they determine best which often can just be it comes in cherry red). When picking between a few models at the same price point you are trading off which parts are going to suck usually. You'll get a decent CPU but you'll have a spinning disk, or IPS screen but integrated GPU etc.
First, who the hell cares? We could demand anything in any business to be honestly named. So what? Marketing will be done, test runs will happen and, worse, the Lead Sales Droid will want it to be sexier. Somehow. Without defining anything at all.
/. question on this silly subject. Do your research (roll versus Intelligence) or fall to the local sales droid. Roll versus wisdom.
Intel could do it, but AMD has repeatedly changed their naming conventions to mimic Intel's in a not very direct way. No, they choose to sound like Intel. While probably in all honesty is trying to tell the customer to buy AMD. No surprise there.
Is ARM going to be forced to start all their CPU designs with WimpyCore? No, and nobody is going to even get past a
Because if you formulated a naming system that more closely represented reality, no mainstream gamer's or enthusiasts (your average poor chap who spends about 1500 on a 'nice' rig) would have bought new Intel processors from the 4700 through the 8700 series.
And the folks building custom server stuff really do their homework, so you could literally call it a banana processor and if they read the specs and it was the best hammer for the job (and fit the socket), they'd buy it.
As for regular consumer products (HP, Dell, etc) home office workstations, laptops, etc most places just advertise the CPU GHz and the amount of RAM, and maybe screen specs if its a laptop. So again you could call it an Apple and people would buy it.
This is a common problem in a number of fields. The usual answer is to turn the calculation around and start with the purpose/function/use-case scenario of the item. In this case, it'd be to sort the list of available items, optimal to sub-optimal, according to performance at compressing video, playing a particular type of game, or whatever. Shoppers can then find the categories that they're interested in.
Debate is a form of harassment. Do not question my truth.
By AMD and Cyrix. Was called P-Rating (P for performance, not Pentium).
It did not work then, it will not work now.
*** Suerte a todos y Feliz dia!
There are a number of issues that make something like this really, really tough to accomplish in practice.
Perhaps the biggest issue is that the product names are selected by the people making the hardware, and we can't really trust them to produce product names which are accurate reflections of their performance, no matter the situation. The best bet would be some sort of independent rating which judged their performance.
Independent ratings are currently done, by a wide variety of organizations. Sites like anandtech.com and tomshardware.com (among others) have pages listing the results of a wide variety of benchmarks for various CPUs and GPUs. But these are often difficult for non-enthusiasts to access, and even harder to understand. I could imagine a "rating system" applied to the hardware which was required to be displayed on the box, so that buyers could see directly.
But then you have the problem (which already exists) that developers will optimize their software stacks specifically for these benchmarks, resulting in overly-optimistic results. GPUs are particularly notorious here, because there is a lot more proprietary software controlled by GPU designers between the hardware and the end-user performance.
Making things even worse, the actual performance of the hardware tends to change over time as software changes. This can be a result of software developers adding new code to make use of new hardware features, or of the hardware vendors updating their own proprietary software.
In the end, getting consistent performance comparisons printed right on the box is a really, really hard problem. It's not necessarily impossible, but it would require a massive investment by an entity dedicated to producing accurate data. It would also require having either the force of law or voluntary agreement by all hardware manufacturers to even get off the ground.
The second you specify how the CPU/GPU performance is measured you will have the vendors gaming the system to get those scores up as high as possible at the expense of actual performance.
You will have them sacrificing needed stuff as much as they can to boost other stuff beyond what is needed to boost that score. Having a CPU that has insane FLOP scores to boost their overall score will come back to bite you in the butt if they sacrificed branch prediction or character operations to do it.
Sorry if my analogy is off, been out of the tech stuff for years but you see the point I am getting at. As soon as you specify how it is measured, you have the vendors working to max out that score at the expense of other stuff. Just like you have cases of GPU drivers found cheating the benchmark software to get higher scores than they really earned.
https://en.wikipedia.org/wiki/...
A table like the above, will explain the breakdown of the numbering. It hasn't changed much with each generation since the core series lineup came in 10 years ago.
Here is a better breakdown with more words than numbers. https://www.intel.com/content/...
But perhaps youre a casual and that's all a bit too esoteric for you?
If you want to easy it up, just go to www.cpubenchmark.net and you can easily compare all cpus and pricepoints. Look at single thread performance if that's all your application can handle (or you are a gamer..), and total performance if its multithreaded. There is a wealth of user submitted data there that i would never view processor advertisements without.
Its really not something you need to spend more than an afternoon getting acquainted with. An exercise that anyone who wants to spend $500+ on a new PC should be more than willing to do. As others have said, basic research is important when buying most things.
-
The average user will have their needs met by any computer built in the past 10 years.
Bullshit. Today a Windows PC with 4GB of RAM (or less) is only useful as a paper weight once you start trying to run a web modern browser and end up paging to a 5400 RPM hard drive.
And it is still accurate.
You are in a twisty maze of processor lines, all alike.
There is a lot of hype here.
This is not specific to CPUs and GPUs, but I am sick of model names/numbers being reused for different products. I was browsing Dell's website recently and found it frustrating to find that the Inspiron 3000 series comes in 15" and New 15" varieties. The 15" variety could use either a Celeron or Pentium processor, while the New 15" ones could be either 7th or 8th generation i3, i5 or i7. Why have three 0s in the number if you are never going to change any of those digits?
And don't get me started on the Inspiron 5000 15" and New 15" range or the Inspiron 7000 15" (but no New 15") range. There is also the Nvidia GTX1060, which comes in two varieties that performs differently.
Refusing to buy a product from a manufacturer that obscures the truth is a rational decision. That's part of why I no longer by anything from Sony. Newer revisions of a product should either have the same features or, possibly, more features. Having later versions have fewer features without warning the customer is ridiculous.
The fact that they felt the need to sabotage already purchased PS3s amongst their other criminal activities just makes it worse.
I won't ever be buying anything from Sony again because of their dishonest behavior. It's unfortunate, that so many people seem to be OK reinforcing those bad behaviors.
It's already bad enough that hardware manufacturers tweak and skew their drivers to eke out another dot at some artificial benchmark program, I don't want them to actually produce their hardware to fit an arbitrary metric that has nothing to do with real world problems because they have to since some illiterates want to compare numbers instead of finding out what they mean.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
The main problem is not that products with different names have different performance - just use a price/performance chart; some websites allow you to sort not just on price but also by passmark score, etc as a customer convenience.
The main issue is that suppliers keep changing the product without changing the product's name or even the SKU. For example when buying two routers from Amazon you might get two routers that look the same but inside they run incompatible chipsets; not just an upgraded CPU or memory or firmware
for average joe, any computer is basically fast and good enough.
On a long enough timeline, the survival rate for everyone drops to zero.
a BMW 320i, Mercedez Benz E300, or a Chrysler 300C? It ain't just processors.
Yours sincerely,
John can-perform-basic-car-repairs sometimes-forgets-anniversaries can't-play-an-instrument digital-logic-design-expert not-very-sporty Smith MEng
Owl tried to think of something wise to say, but couldn't.
WARNING: Measurements are approximate.
A line of CPUs having consistent numbering. Maybe generation-cores-clockspeed or something. It would at least allow for easier comparisons within a line of CPUs. Might should add cache amount, but within a line that doesn't vary as much. Letter or number designation would be fine (e.g. Q for quadcore H for hexacore). How long has i3-i5-i7 been around and look at the names? AMD is similar. Similar for GPU maybe with amount of ram in there.
The model number is cooked up by Marketing or, at best, some accounting system. It has even less significance than a Mercedes car model number (which at one time represented the size of the engine to some degree of approximation, but no more). If you want more information about the part, you have to read the specs. Intel and AMD both provide handy comparison features at their web sites.
No, you don't need manufacturers to agree to anything, and in fact you ought to be ignoring whatever it is that they say they'd like.
You just need someone else (not the manufacturers) to associate benchmark figures with the part numbers. And there can be as many different "someone else"s as there are types of users.
Let the part numbers be arbitrary. They're just names, like "corvette" or "civic." If today your opinion is that foomark is the best measure of "real world" performance, then just
select cpus.* from cpus
join benchmark_tests on benchmark_tests.cpu_id=cpus.id
join benchmarks on benchmarks.id=benchmark_tests.benchmark_id
where benchmarks.name='foomark'
order by benchmark_tests.result desc limit 10
With my son going off to college, I wanted to buy him a laptop capable of running 3ds Max so I've had to do a lot of comparisons. My complaint is the Intel Core model numbers don't describe even the basic characteristics of the processors. e.g. while the i7 processors typically have 4 cores with hyperthreading, the 7nnnU and 7Ynn processors only have 2 cores (with hyperthreading). Then there are the four Kaby Lake Refresh "8th generation" processors which all have 4 cores (with hyperthreading), but much lower normal clock speeds and very high (but largely irrelevant) turbo clock speeds.
cpubenchmark.net helps, but also https://en.wikipedia.org/wiki/Intel_Core
I was on the team for Intel Labs Europe/Toshiba Europe when they decided to take the MHz rating off of new CPUs to make it "easier for customers" who "don't want to read numbers." I told them from the very start that it was a bad idea that was clearly done for marketing, not user friendliness, but they told me to shut the fuck up and write it up as though it was the greatest thing since sliced bread. Fucking idiots.
- In Soviet Korea, only old people loose all their bases to Natalie Portman's petrified hot grits overlords.
Because processors in general are general purpose in nature, it would be impossible to assign a model number based on some made up score. How many processor cores, how much L1, L2, L3 cache is there, clock speeds, how well optimized the operating system is for the chip in question, chipsets, RAM speeds, storage speeds.
Special purpose chips, not having flexibility in terms of what gets run, does not have that sort of confusion. How quickly can you handle various video codecs for example, will not have as many outside factors when it comes to the performance.
One thing that SHOULD be standardized is within each company, what the core count is, and if SMT is enabled/available or not. The Intel chips for example, some are dual-core, some quad-core, some have SMT or HyperThreading support, some do not. AMD also has some confusion, all AMD A8, A10, and A12 chips are quad-core for the CPU, the A9 is a dual-core chip. There are some A4 and A6 chips that are quad-core, while most are dual-core. Trying to navigate that sort of thing does take an online search to see just how many cores and threads a given processor has. Adding a U at the end for those very low power mobile chips can take a quad-core for the regular version to a dual-core for the "U" version, but not always.
If you buy a Lintel Gore j6 123456 AZ microprocessor with a 3.333 GHz clock then it should not only offer the same real-world performance as any other microprocessor with the same brand name, part number and clock frequency, but it should even be fully hard- and software compatible. In practice, you need to check the chip 'stepping' number to see if two parts will function identically or only roughly the same.
Caveat Emptor
real world figures are an anathema to chip flippers. If chip flip A came out with a chip with a certain number chip flip B would beg, borrow and/or steal to top it so the obfuscate things. Besides, it doesn't matter, if we are talking pc then there are only 2 real chip pushers and they work together most of time to manage the market.
Can you only understand sentences with 1-syllable words? Do you only read the "simple English" version of Wikipedia??
There's no such thing as "real world experience" in performance of most anything. The world is far more complex than that and so are man's engineered solutions to problems posed or created by the complexity. There is no simple answer that can be encoded into a product number and if you think there is or should be, your IQ is not high enough to be capable of specifying or choosing such things. Let people with more brains do that for you.
Given the end of the PC era and even the end of Moore's Law this already become somewhat moot - mobile already defines computing for the majority of the world's economies. Microsoft is putting Windows on the back-burner. Apple is eliminating Intel as its Mac processor and replacing it with its own derived from its mobile products. Intel is staggering right now seeing to reinvent itself without any success so far.
Average Joe user doesn't care about any of that stuff - they just want to use Microsoft Excel, Outlook and Word for work stuff and a browser for the intertubes.
If you're wanting to compare parts then you're going to have to spend time researching on review sites like AnandTech and Phoronix and come up with a list of pros and cons to inform your purchase.
The performance of a computer is a multi-dimensional thing, and it's impossible to produce any single-dimensional measure which can reflect it.
Even car buyers understand this sort of thing :-) There are plenty of numbers available for each model: peak power, peak torque, fuel economy, and so on. But even simple things like fuel economy isn't single-dimensional: they have to quote highway and city values. Then there are the data that aren't direct measures, but are simple specs: number of cylinders (analogous to the number of cores), cylinder bore and stroke, and so forth.
It's a foolish quest to try to reduce performance to a single number. You have to consider a multitude of factors.
If this is a consumer thing, why not index on the biggest raw costs to what make the chip suitable to it's purpose.
Transistor density, clock speed, and maybe R&D costs.
120 characters ought to be enough for anyone