'Faster than' means

[Eolirin]

you have results for 4k, I wouldn't call 36% (down from 50%) perfectly scaling for 60

Idk, math is kind of weird on that.

If we use 60 as our baseline instead of 40, then 40 is ~66% of 60, and it's ~73% of the speed. The scaling is pretty close to what it should be looking at it like that.

But I'm honestly not sure which way is the right way to do that comparison.

 

[snc]

Idk, math is kind of weird on that.

If we use 60 as our baseline instead of 40, then 40 is ~66% of 60, and it's ~73% of the speed. The scaling is pretty close to what it should be looking at it like that.

40 is base 60/40 = 1.5x and real results we got is 1.36x

 

[Eolirin]

40 is base 60/40 = 1.5x and real results we got is 1.36x

Sure, but why does 40 need to be the baseline, and how is the discrepancy between taking the exact same numbers and going in the other direction resolved?

 

[snc]

Sure, but why does 40 need to be the baseline, and how is the discrepancy between taking the exact same numbers and going in the other direction resolved?

to know how much something is faster than other we take slower one as base, just math

 

[Eolirin]

to know how much something is faster than other we take slower one as base, just math

That's not true, it should always work in both directions, but I think I resolved it anyway. The gap is actually the same doing it both ways, it just seems smaller at first glance doing it one way, but correcting for the difference in scale it's the same.

 

[snc]

That's not true, it should always work in both directions, but I think I resolved it anyway. The gap is actually the same doing it both ways, it just seems smaller at first glance doing it one way, but correcting for the difference in scale it's the same.

the other way could be confusing for example something is 2x faster, so slower as base we got 2/1 and we got simple 2x, faster as base and we get 1/2 = 0.5 and somebody wrongly could assume its only 50% difference while its 100%

 

[iroboto]

the other way could be confusing for example something is 2x faster, so slower as base we got 2/1 and we got simple 2x, faster as base and we get 1/2 = 0.5 and somebody wrongly could assume its only 50% difference while its 100%

No, technically someone should see it about 50% as fast. There's not reason that someone should assume the difference is 100%
40 out of 60 is a glass that is 2/3 full. There is nothing wrong with that fraction.
And 60 out of 40 is a glass and 1/2 full.

It just depends on which number you want to make the size of the glass.

If you have an option to make a scale out of a dataset. Do you select your maximum number using the smallest number? Probably not. So it depends on what you're trying to achieve. Most people will explore the min/max of a data set, and create a logical scale from there.

It's not wrong for someone to look at the RDNA 2 series and see it goes from likely 36CU all the way to 80CU.

And make 80CU the size of the glass so that everything fits neatly into it as a working fraction.

Being able to select the correct scale and fractions is really just a representation of the story you are trying to tell:

• If you want to look at the whole family of devices, one may desire the baseline to be 80CU.
• If you want to look at just 1 device, you probably want to baseline relative to that device.

With respect to the debate you guys are having, I don't know which way I'd go. I think looking at a family of devices and seeing a scale, I think I'd use 80 CU as the glass and look at performance relative to that if the goal is to look at CU scaling.

 

[snc]

No, technically someone should see it about 50% as fast. There's not reason that someone should assume the difference is 100%
40 out of 60 is a glass that is 2/3 full. There is nothing wrong with that fraction.
And 60 out of 40 is a glass and 1/2 full.

It just depends on which number you want to make the size of the glass.

If you have an option to make a scale out of a dataset. Do you select your maximum number using the smallest number? Probably not. So it depends on what you're trying to achieve. Most people will explore the min/max of a data set, and create a logical scale from there.

It's not wrong for someone to look at the RDNA 2 series and see it goes from likely 36CU all the way to 80CU.

And make 80CU the size of the glass so that everything fits neatly into it as a working fraction.

Being able to select the correct scale and fractions is really just a representation of the story you are trying to tell.

guarantee you that many are confused when faster is base,edit: your comment about 50% was about qouted example of 2x faster ? if so then you are wrong ;d something 2x faster is not 50% faster but 100% faster

 

[see colon]

guarantee you that many are confused when faster is base,edit: your comment about 50% was about qouted example of 2x faster ? if so then you are wrong ;d something 2x faster is not 50% faster but 100% faster

What? Two times faster is 200% faster. Twice as fast is 100% faster.

 

[snc]

What? Two times faster is 200% faster. Twice as fast is 100% faster.

2x is 100% faster, English is not my native so I don't know if 2x can be translated to two times faster or there is difference and has to be translated to twice as fast

 

[zed]

What? Two times faster is 200% faster. Twice as fast is 100% faster.

huh, I would say 2x faster = 100% faster
but then again I do suck in english, and its the language Im most proficient in

 

[Pete]

Can we all disagree that it’s:

y faster = x + yx.

y as fast = yx.

{Two times = 2x = 2* = 200%} faster = {300% = three times} as fast.
{One times = 1x = 1* = 100%} faster = {200% = two times = twice} as fast.
{Half times = .5x = .5* = 50%} faster = {150% = one and a half times = half again} as fast.

I like using x as the implied standard variable name as it makes this 2x as confusing, aka 100% confusinger. Your guess is as good as mine as to why I used {} rather than [].

 

[Metal_Spirit]

Can we all disagree that it’s:

y faster = x + yx.

y as fast = yx.

{Two times = 2x = 2* = 200%} faster = {300% = three times} as fast.
{One times = 1x = 1* = 100%} faster = {200% = two times = twice} as fast.
{Half times = .5x = .5* = 50%} faster = {150% = one and a half times = half again} as fast.

I like using x as the implied standard variable name as it makes this 2x as confusing, aka 100% confusinger. Your guess is as good as mine as to why I used {} rather than [].

Friends.. Native language doesn´t matter... this is math. And yes.. it can be confusing.

100% faster is twice the speed!

If my car goes to 100 MPH, and my other car is 10% faster, he goes to 110 MPH... If it is 100% faster, he goes at 200 mph...

Question is... 110 MPH = 1,1*100 MHP, and 200 MPH is 2x100MPH, and that means 110 MPH is 100% the original speed+10% the original speed=110%, and 200 MHP is 100% the original speen+100% the original speed = 200%.

So both are correct... The wording faster removes the need of the use of 200%, since we are adding over the 100%. So 100% faster is the same as saying that in comparison this car gives 200% the speed of that other.

 

[see colon]

Friends.. Native language doesn´t matter... this is math.

The language matters I think, because the nuance is in the use of the words "faster" and "as fast". Those words have definitions close to each other but one implies inclusion of the original value while another does not.

 

[zed]

I do agree with you see colon that language matters, I tried to have a look online for authoritative (i.e. not someone on a forum) definitions of the language but couldnt find anything! Maybe someone elses google-fu is better

 

[dobwal]

What? Two times faster is 200% faster. Twice as fast is 100% faster.

Two times faster and twice as fast is the same thing.

100 mph X 2 = equals twice as fast or 2X faster.

100% faster is quickly calculated by taking the base and multiplying by 2. 100% indicates the size of the increase without accounting for the base itself.

In other words 100% is literally equal to 1 (conversion of percentage to decimal) so any number multiplied by 100% is equal to itself.

100% increase (or faster) would be calculated as 100 mph X 2 or 100 mph (base number) + 100 mph (the increase in speed) = 200 mph.

 

[zed]

https://www.nytimes.com/2007/10/28/opinion/28iht-edfreeman.1.8081659.html

It's those commentators who are confused, say the Merriam-Webster editors.

"Times has now been used in such constructions for about 300 years, and there is no evidence to suggest that it has ever been misunderstood."

• Thanks for reading The Times.

Subscribe to The Times
They also dismiss the related idea, often invoked in newsrooms, that "five times more" isn't synonymous with "five times as much," but really means "six times as much": "It is, in fact, possible to misunderstand times more in this way, but it takes a good deal of effort."

I've heard people argue that threefold and its brothers are ambiguous, too, but I doubt it. The Oxford English Dictionary says threefold has meant "three times as much" since it was an Old English word; it seems perverse to keep looking for ways to make it mysterious.

 

[see colon]

Two times faster and twice as fast is the same thing.

100 mph X 2 = equals twice as fast or 2X faster.

100% faster is quickly calculated by taking the base and multiplying by 2. 100% indicates the size of the increase without accounting for the base itself.

In other words 100% is literally equal to 1 (conversion of percentage to decimal) so any number multiplied by 100% is equal to itself.

100% increase (or faster) would be calculated as 100 mph X 2 or 100 mph (base number) + 100 mph (the increase in speed) = 200 mph.

What? No, two times faster is a 200% increase. Twice as fast is a 100% increase. The word "faster" is a rate of increase, and the base value must be added to it. "Two times faster" is descriptive of the rate of increase only, and not the base so the base must be added to it to calculate the final number. So base times rate of increase to find thee rate of increase, added to base to find the total number. "As fast" includes the base value and is comparative to that base value. So base multiplied times the comparative value (2 in the case of twice). You say it right there at the end. "100% increase (or faster) would be calculated as 100 mph X 2 or 100 mph (base number) + 100 mph (the increase in speed) = 200 mph". That's 100% (or 1 times) faster, or twice as fast. 200% faster would be two times faster.

-Edit-
After putting some thought into it, I think "two times faster" may be grammatically incorrect anyway. It's like saying "more faster".

 

[snc]

lmao with confusing of percentage usage ;d its really better just to keep to multiplayers like 1.5x, 2x and don't use percentage at all (you can't interprent 1.5x or 2x in different ways) ;d and to do that slower should be base

 

[see colon]

lmao with confusing of percentage usage ;d its really better just to keep to multiplayers like 1.5x, 2x and don't use percentage at all (you can't interprent 1.5x or 2x in different ways) ;d and to do that slower should be base

What if we multiply by .5?