980 ti amp extreme for 340 or a 1070 for 380

Status
Not open for further replies.
Solution
1. Resolution ?

2. 1070 is 14% faster

3. 980 Ti (30+%) overclocks a lot better than the 1070 (18%) which leaves the 1070 w/ about a 3% edge
in performance.

4. The 1070 is more power efficient, quieter and throws less heat.

5. is it worth paying 29% more for a 3% performance advantage, lower heat, noise and power ? ... that one only you can decide :)
1. Resolution ?

2. 1070 is 14% faster

3. 980 Ti (30+%) overclocks a lot better than the 1070 (18%) which leaves the 1070 w/ about a 3% edge
in performance.

4. The 1070 is more power efficient, quieter and throws less heat.

5. is it worth paying 29% more for a 3% performance advantage, lower heat, noise and power ? ... that one only you can decide :)
 
Solution

Tradesman1

Legenda in Aeternum


_____________

+1 the other percentages cited are questionable also i.e. it looks to assume the OCability of ALL 980Tis and ALL 1070s are the same and it varies greatly depending on the individual GPU chip primarily but have to fator in the design of the card and an individual's OC ability. Many see/read a review and think it's true of ALL cards. We've seen people fry components because they read a review and simply try whatever settings were used in a review
 


apples and apples comparison. What I didn't do is compare a budget model with a high end model

The amp extreme is one of the highest end cards there is ... would logically be compared with the EVGA Classified, Asus Matrix, etc.

So what I didn't do was compare one pf the best performing 980 Tis ever made with the cheapest available reference design 1070 as that would be far from "apples and apples".

So... what i did was just compare with the "gaming cards"

Strix is $460 - http://www.newegg.com/Product/Product.aspx?Item=N82E16814126109
Gaming X is $450 - http://www.newegg.com/Product/Product.aspx?Item=N82E16814127947

$450 / $340 = 32% more expensive

However, with the $470 1070 AMP Extreme now out, we are in a position to compare apples and apples.

http://www.newegg.com/Product/Product.aspx?Item=9SIA2F84889680

$470 / $340 = 38%



 
He is saying he can get a 980ti for 340 or a 1070 for 380 so thats only 40 difference thats not 29% its 11%. Also considering the 1070 beats out a titanX maxwell (which performed better than a 980ti) i'd say for the 40 extra currency the 1070 is ABSOLUTELY worth the extra coin. More performance, less heat, better power efficiency, and the 1070 overclocks amazingly. over 2ghz. I'd say thats fantastic overclocking.
 


1. The cost differences are by no means questionable unless you can't do math or want to purposely mislead by comparing apples and grapefruits. When comparing two automobiles, you don't compare two in the same model line when one as large tires, chrome wheels, V8, and a host of other features with the "economy model"....

But I was wring, the actual cost difference "apples and apples" is 38% based upon the links provided

2. Of course, there's no data to support this questionable claim but if you bother to do the research

TPU results for example ... +2 / -1% of average is a pretty tight range

Gigabyte G1 - 131.4%
Palit Jetstream - 129.7%
Asus Strix - 128.4%
MSI Gaming - 127.2%
Zotac Amp - 127.1%

Factor in all the other reviews as well as

our own builds ... and the range is rather small.

2. It has been argued vehemently on these very forums my many the 1070 AIB cards have no difference in performance. I have seen a maximum variance of 4%

Yes all cards vary but we have to remember that Boost 3 brings a whole new ballgame. The cards are physically capable of more performance ... this has been mentioned in many reviews, but because of Boost 3 and BIOS limitations, we are just not seeing a wide variance. There are certainly no 1070s getting up in the high 120s / low 130s OC percentages. If you want to claim otherwise, let's see the data
 

Tradesman1

Legenda in Aeternum


______________________

You and I read the post ;) Which is always helpful when responding - that way we know what we are responding to ;) Same on the AMP - Zotac has I think 3 (maybe 4) 980ti AMP models the Omega and Extreme on the upper side, then their regular 980TI and an Artic Storm model
 

Tradesman1

Legenda in Aeternum

______________________

Yes the prices are questionable - where you come up with yours (from whatever else you have been googling/reading) obviously aren't the prices in the opening post of the thread. Try reading the opening thread rather than blabbering along.
 
also the amp is not the best out there. The amp extreme is what is compared to a classified, gaming z, lightning, matrix etc. The amp edition is no more than a regular card with a slight overclock.

You seem to be confused. My link was clearly to the AMP Extreme

The OP clearly stated the amp Extreme
shall i save 40 bucks and get the amp extreme 980ti

What is your issue here ?

BTW, that regular card did pretty good in testing ... it actually outperformed the AMP Extreme in TPUs test. It also prices a point I will make later on ... clocks means beans, you can get better performance with lower OCs. That's the way boost works sometimes

ZOTAC GTX 980 Ti AMP Extreme = 1505 MHz / 2096 MHz / 129.1 fps
ZOTAC GTX 980 Ti AMP! = 1465 MHz / 1990 MHz / 130.4 fps




Geez .... where to start.

1. Be aware that highest cost doesn't mean highest performance. Knowledgeable gamers weren't buying Titans, they were buying 980 Ti's since they were much faster. The 980 Ti toasted the Titan X in gaming because a) the Titan X was only a reference card and b) the 980 Ti overclocked so much better. That's a 100% to 89% (12.4%) win... that 980 Ti card OCd 10.6% further...the Titan X being reference only OC'd 4%

perfrel_2560_1440.png


https://www.techpowerup.com/reviews/Gigabyte/GTX_980_Ti_XtremeGaming/26.html
https://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_Titan_X/32.html

112.4 x (110.6 / 104.0) = 119.5 win for the 980 Ti

With both cards overclocked, the 980 Ti was therefore just a hair under 20% faster than the Titan X.

2. Show me this amazing OC... and I mean show me actual performance for THIS particular $380 1070. The cheapest 1070 on pcpartpicker is a mini, which will be limited in performance just like all reference 1070s are limited by throttling. 2000MHz OC... big whoop ! ... as we so so well above comparing the Zotac cards, more Mhz in and of itself does not mean better performance.

Who cares what the numbers are if they don't deliver anything ? You have two problems here:

a) As so plainly demonstrated above, more MHz does not equate to more fps.
b) The bargain basement reference 1070 will throttle and can't hold that OC

When I indicated % advantage when OC's, I am only using actual results, that is in how much more fps.

http://videocardz.com/60838/msi-geforce-gtx-1080-gaming-x-is-much-better-than-founders-edition

GTX-1080-FE-clocks-over-time.png


There's your $380 1070... starting out at 1880, but dropping to 1680 .... so much for that "amazing" OC.

3. Let's look at the actual cards in question

133.7 fps (133.7 Reference OC / 117.9 = 13.4% increase in fps OC'd)
https://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_1070/27.html

129.1 fps (129.1 AIB OC'd / 102.0 reference stock) 26.5% increase in fps
https://www.techpowerup.com/reviews/Gigabyte/GTX_980_Ti_XtremeGaming/26.html

133.7 fps / 129.1 fps = 3.56% faster ... what a coincidence !

4. As for less heat, better efficiency ... these are secondary considerations and do not have equal importance to all users. I do not wish to impose my view of the importance of these issues on another user. For example the fact that the 1060 has better efficiency than the 480 @ $5.70 in electric costs per year will be of no concern to most users.

It's also produces more heat and is louder, but these are two advantages many would easily give up if the 480 was faster... but it isn't. I have oft been told, "I don't care if its louder cause I play w/ headphones on". So while I did very clearly state that the 1070 is cooler, quieter and slightly faster. I did it by presenting the facts and let the OP to decide how much weight to apply to each one.

5. Simply put, it's a value comparison ... the difference between a high end AMP extreme or a bargain basement 1070 is not apples and apples.

a) You can not put the high end AMP Extreme on the same level as a bottom of the barrel reference 1070. This is a false equivalency. It's like comparing 2016 Brand A Pickup Truck w/ V8, 4 WD, all all the options with a bare bones 6 cyl, 2WD 2017 Brand B Pickup Truck

b) The real measured performance difference between the cards when both are overclocked is very small.... 3-4%

c) The 1070s advantages in heat, power efficiency and noise are real, but the importance to each user will vary widely.

d) The feature, component, cooling, overclocking differences between the cards are real, but the importance to each user will vary widely. If you don't overclock for example, many of these advantages will not be significant factors in the decision making process.

If it were a personal build, or one for one of our users, I would nix both choices. The logical and equivalent comparisons would be between:

-cheapest 980 Ti you can find versus cheapest 1070 you could find

-lowest cost of quality AIB 980 Ti w/ desired features you want (backplate, cooler design, etc.) that you can find versus cost of quality AIB 1070 you could find

With the OPs proposed choice, the only thing I can and did recommend is to weigh the relative value of what is on the table without ignoring anything. To get a 1070 with the equivalent feature set as the 980 Ti being used as the base reference, you would have to pay more than 30% more. The **value** difference is more than the $40 cost difference.... just like when you trade in a car. A year / model being traded in may have a base price... but then adjustments in value (positive and negative) must be added when you go thru the entire package...No AC ... minus $230 ...smaller engine ... - $500. Same thing here. The reference card throttles, AMP Extreme doesn't... how much is that worth ? Don't have a backplate, how much is that worth ? Apples and apples.
 
you seem to be confused and are adding in what you think vs what the OP is talking about. He found a 980ti and a 1070 he is looking at specifically apparently. Also The 1070's and 1080's only throttled when using the stock fan profile which capped at i believe like 60% fan speed. Every tester including Jayztwocents showed once you bump the fan profile up the cards had no problems holding their clocks. https://www.youtube.com/watch?v=OXUo1S55ZUM Again even the stock 1070 (reference) Would be a better choice for $40 more over the 980ti anyway you want to slice it period. Argue all you want, the 980ti performs less overall, uses more power, is a older card with older architecture. You agree'd with what me and tradesman1 have been saying. The 1070 is better in every single way and for only $40 more. You also have no clue what 1070 the OP is talking about as he could be looking to pickup a used one possibly for that $380 he is talking about. You are making assumptions and trying to justify based on info you do not have. Even the lowest end 1070 will all around be better and faster no matter how small that amount is and will be all that for only the $40 difference. Hell the electricity costs he would save over a years time would pay for that extra $40 he spent getting the 1070......maybe even in 6 months.


 
What you are trying to do by presenting a 980ti amp extreme to a 1070 amp extreme is understandable but the OP is asking should he get a 980ti for $340 (regardless of what model it is because even the lowest end 1070 is still faster) or a 1070 for $380 and is the difference in cost worth it. It's a fact that for that $40 you get ALOT in terms of overall features and performance even at the lowest end of the 1070 spectrum. Is the $40 worth spending for the 1070? Only the OP can decide that but do not confuse the OP..It's a simple question that needs a simple answer. Power saving, higher clocks/performance, newer technology and offers as stated energy savings that alone would pay for the difference in at most a years time. Those are facts of that card and how its better over the 980ti at even the highest end level. As you get higher up in the 980ti line (amp extreme) you are using considerably more electricity, creating more heat inside your case and is a older tech card. Now those things dont mean the 980ti is a bad product it just means that for $40 to get to a 1070 is arguably more than worth it.
 

Tradesman1

Legenda in Aeternum
azzazel_99

and argue he will, this happens all the time. The other day he told an OP a mobo was the best rated of all at Newegg (it had 2 reviews both for 4 stars), he was thinking a completely different mobo - so then the response was it's in the same family of boards..... There is no winning when one like this is completely changing the subject, and throwing a lot on non-related info as we've seen here.

Best to walk away and ignore the blabbering, as it will continue. ;)
 


It's called research. Relying on actual published testing and advertised costs for equivalent featured products. You might try, at last once, posting a single piece of data to support your position. It's also called avoiding a false equivalency... I doubt you'll look as learning is an anathema but here's a link

https://en.wikipedia.org/wiki/False_equivalence

You can not compare a very specific 980 Ti to "any 1070". You can't make a "equivalent comparison" between two items that have different features and here we have a widely varying feature set. To make a legitimate comparison you must get down to a more basic "980 Ti versus 1070 equivalent comparison". The only way to do that is to compare cards where both are equipped with equivalent components and features so you can isolate the performance and cost difference of the GPU difference ONLY.

My costs came from newegg for equivalently featured cards....after eliminating the false equivalency. In other words, and as so many times repeated, an "apples and apples comparison". It's a value comparison. When you compare two different items, an accounting has to be made of the relative value of the items included.

I am making an assessment of worth.... here's what an apples and apples comparison looks like:

-The 980 Ti Amp Extreme is $340 ... this is the basis of the OP's choice. It's named, it can be quantified. The unnamed 1070 can not ... we do not know what it is so it can **not** be quantified
-The 1070 Amp extreme is $475

Now for non-equivalent comparison. The $475 AMP extreme $65 more on newegg than the cheapest $410 1070. Why ? Because they are not equivalent. What is being given up ? How much are those "given up" things worth ? To compare the relative value or A versus B when A and B are no where near equal, you must make a value assessment of those differences.... right now the value of the AMP Extreme 1070 over the cheapest 1070 on newegg is clearly $65. That's not blabbering, that's a documented fact.

So, yes, comparing a 980 Ti AMP Extreme to a low budget 1070 is a false equivalency. And when you have a false equivalency, just like Kelly's Blue Book, you have to work from a "base value" and make $$ adjustments for each of the items that are not equivalent.

if one wants to find a 1070 equivalent to the greatest extent possible to the 980 Ti AMP Extreme, it will not cost $40 extra, it will cost $135 extra and we know that because we can look it up. So what adjustments have to be made?

... Is this 1070 a mini ? (closet in price to $380 is the $385 mini). Do I want to accept the limitations of a mini ? How much is that worth ?

.... Am I giving up a backplate, I could buy one for $20-$30 if i want it ... is it worth getting ? The answer to that question will vary among individuals but I'd say 65% won't care. If ya want one, than ya gotta add $20-30

... I am giving up a card that heavily overclocks to get one that throttles even at stock settings... The answer to that question will vary among individuals but to most I think this will be a significant concern. For me, I don't like the idea of owning a card that will throttle every time I play game. I will never own one.

I can not speak to the OPs willingness to give up certain features; as clearly stated, whether these are given up and what is the worth of each item given up will vary by individual. I can not make an assessment for the OP, but what I can do is indicate what an equivalently featured 1070 would cost, that's what I said I was going to do and that's what I did.
 

Tradesman1

Legenda in Aeternum
Your statement

"5. is it worth paying 29% more for a 3% performance advantage, lower heat, noise and power ? ... that one only you can decide "

A $340 card or a $380 card - there is no 29% difference, You can rationalize all you want, you can blabber all you want about stuff no one wants to hear - this is simply you insisting anything you say is right... Same as you did on that MSI mobo 'Duh, well it's the same family" , no offense but being in the same family does not make it 'best rated'. You do this all the time, to try and prove you are right or think you are smart. It gets rather ridiculous.
 
I compare the highest end 980ti to eh lowest end 1070 to prove that the BEST 980ti has to offer is still not enough to beat out the lowest end 1070 lol. That means that at any level you look at the 1070 beats the 980ti at any level you look at. The OP is looking at a $340 980ti and a $380 1070. So for $40 you get a better faster card period end of story. You can't argue with that. Its a fact. The 1070 is better all around and faster........end of story. Lol i dont have to post proof because there are plenty of that stuff on the web a simple good search would show lol. So if OP an get a 1070 for $380 i would sure as hell do it.
 

Tradesman1

Legenda in Aeternum

______________

+1 the 1070 is well worth it. And I'll be leaving this thread, if I stay, I can see this falling into a full blown argument, where, as I do have ethics and a fair amount of integrity I would be forced to ban both Jack and myself, as I would be wrong, and I openly admit when I'm wrong :)

My apologies to NeymarX - if I can be of help feel free to send me a PM
 


Oh boy the inevitable "Period" ... in other words ... you won't listen to any facts to the contrary and will create false strawmen arguing against positions you created rather than those actually put forth.

1. Quoting JayZ is not exactly enhancing your case. It's the tech equivalent of quoting Sarah Palin. Lets remember that this is the same guy who is still ranting that 3 and 4 way SLI is still "officially supported" by nVidia and the same guy who thought drilling thru his X99 MoBo was a good idea. At two cents , he's grossly overcharging.

2. You keep choosing only factors that support your position and ignoring those that don't ... does the 1070 have a backplate ? Is this important to OP ? I don't know, you apparently do. Again, lets go back to the truck analogy, you are focusing on mpg, driving range of the newer model with the smaller engine but keep ignoring torque, towing capacity, 4 WD vs 2WD etc. As the OP has not stated the relative importance of various features, I think what is important is ... not what you think is important ... but what the OP thinks is important. I don't know what that is, you apparently feel you do. This is a simple concept, the relative value of two choices is not only what you get for the extra money but what you give up. We both have no idea which items are important on either side; I am unwilling to make a judgement either way w/o this knowledge, yet you have no qualms abut doing so.

3. I can't make decision for the OP, I can only supply facts. I didn't tell, OP what to buy. The fact is the 1080 AMP Extreme costs $65 more than the cheapest 1070 on newegg, people are buying them so the cost difference is clearly not considered outrageous. You can argue all you want that those $65 worth of differences are not important ... to you ... but a) other folk obviously do as they are being sold and b) obviously you aren't the OP.

4. I don't know the OPs knowledge level; I don't know the OPs desired features. In the absence of that information, unlike you, I don't fell that I am in a position to make decisions for him. You are arguing against yourself. I can't tell him what to buy because I don't know what 1070 is being considered. Yet, while claiming that this is a weakness in my position, I didn't make a recommendation either way... you did.

Your claim is a complete falsehood, I never said "get a 980 Ti". All I did was point out what a 1070 that was equivalent to the 980 Ti AMP Extreme would cost and suggested that the differences between the AMP Extreme and a $380 1070 be considered in making the decision. Why do you object so strongly to the OP making those evaluations and making up their own mind ?

5. Let's even ignore that the OPs two indicted choices are apples and grapefruits ....

11.7 % increase in cost for a 3-4% performance improvement is not exactly a positive return on investment.

Personally, I would chose a 1070 over a 980 Ti ... however that's based upon my goals and criteria that is important to me. However, I won't impose my views on the OP.

As I already said, sound is very important to me, it affects my choice. But I won't tell a poster that he should choose this or that because card A is quieter than card B, I will just say that the typical low budget 1070 is 36 dbA and the Amp extreme is 35 dbA and let the poster make his own decision. Not sure why you need to dictate a choice because a feature is important to you, everyone must be of like mind.

You keep repeating this over and over using a strawman that has nothing to do with what I said but which you created. The OP didn't name the 1070 in question. The only apples and apples comparison that I could make was therefore was to compare it with something equivalent to the specific 980 Ti which was named. To my eyes, the features that cards like the AMP Extreme have value; the marketplace clearly agrees but you don't. OK ... but the OP, being a rational being I assume, once presented with the data at hand is capable of making a decision that suits the things that he / she considers most important. You are working extremely hard to press an argument that what is important to you is the only thing that matters. I am completely content with him / her evaluating those differences, assessing the relative costs and making a decision based upon items that are important to and costs associated therewith.

If OP comes back with a specific model, then I can make a detailed comparison but again, all I can do is place the features and data for comparison... only the OP can decide if it's worth the difference in cost.
 


Actually the statement was much longer than that ... so lets get back in context shall we. The context was comparing "apples and apples"

The 3% was taken from two hi end AIB cards ... therefore the cost difference was taken from hi end AIB cards. As for the the personal Trump like attacks.

I always cheer up immensely if an attack is particularly wounding because I think, well, if they attack one personally, it means they have not a single political argument left.... Margaret Thatcher
Read more at: http://www.brainyquote.com/quotes/quotes/m/margaretth114262.html



 

Tradesman1

Legenda in Aeternum
azzazel_99,

Would suggest walking away from the fruit cart also (Apples, oranges, and now grapefruit????) it will continue to deteriorate (will prob get into planes, trains and automobiles next) and the responses will get longer and longer and not be saying anything
 
Status
Not open for further replies.