dexT

19 minutes ago, mr moose said:

 

 Let's assume it is already for mass release as a gaming GPU.  It would be on time from a consumer perspective, but early from nvidia's financial perspective (they already have the top cards and they are still selling).

 

 

And that brings us back to the sandbagging.

780 TI was the fastest GPU when the 980 came out. The performance gap is much smaller between Vega and Pascal than Vishera and Kaby Lake was on the CPU side and I'm sure you know about the hate Intel has been getting lately for the same thing.

51 minutes ago, mr moose said:

 

So the question your really asking is why wouldn't he consumer be happy if they released Volta now fully working, faster and the same price.  The answer is the consumer would be happy, but for now only.  As any practice that reduces revenue effectively reduces R+D expenditure,  the end result is a poor performing product.  So yes, you might get this one early and be happy, but at the expense that the next one will not be as good, and the one after that and the one after that.  Just ask AMD how hard it is to produce a leading GPU without any R+D cash.

It's not exactly early though is it? Unless this card is out by January it will be the biggest gap since 280 to 480.

18 minutes ago, Dylanc1500 said:

I want just interject a thought for you. You are making the assumption that because the V100 (which, mind you costs roughly $1000 to produce) die is complete that they will instantly have something compelling enough for release to the consumers. They very well may not have consumer "purposed" dies ready. There is a lot of relevant information that is not available. If they jump the gun before it's a fully compelling lineup, then they might not make quite the revenue as compared to if they waited till they could develop it a little further to make it more enticing for people to purchase. It is a game of timing. To early you lose money because people won't buy the product because it isn't enough of an upgrade to purchase, wait too long and someone my purchase the competitors product or purchasing slows to a point that they aren't bringing in the revenue that is needed. It's a balance.

 

but that's all I'm going to say, this is kind of off topic and it is bed time for me lol. 

 

 

Good post and I understand, but it would be better for the consumer(what the argument is) to not "delay gratification" and bust out the better, faster cards for the same price. There are repercussions like you state, oc.

6 minutes ago, mr moose said:

But they didn't want to rush it, that's the point.  they had to answer Ryzen, if there was no Ryzen then CL would have been held of until they had supplies to meet demand. and also they wouldn't have had to do the 7700K bundle deal. If they had of pushed CL out the door earlier then they may not have had anything ready to counter Ryzen.  And that hurts sales.

 

 Volta being ready for release and being prudent to release early are two different things,  Because we don't know what comes after volta nor how much money Nvidia will need to finish it.    As I said before, sending volta out early may well hurt sales of their 10 series which will reduce revenue, if that revenue is needed for the R+D of what comes after volta and there isn't enough, then that will either delay said release or said release will have less features/more problems. neither options are beneficial for consumers. 

 

 

 

So you want them to wait until sales for pascal is zero? They could have not cut the last round pascal dies and launched Volta... they chose to make more pascal when Volta is done.

5 minutes ago, mr moose said:

  Imagine what would have happened to Intel's sales if they couldn't bring CL release forward. 

They rushed it to market not delaying the customers gratification.

4 minutes ago, mr moose said:

 

Just using the figures for an illustrative purpose.   My point is they would sit on it long enough to maximise profit and make sure the next product was better, rather than just release straight up and hope it didn't have a negative effect on sales and leave them with nothing to release later (or worse an update fix a rushed launch).

They chose the move and chose to cut out more dies. The product is ready. The next generation after Volta would come sooner for the consumer the sooner Volta is released which would be a good thing for the consumer(faster and more features for the same price is better for the end user).

5 minutes ago, mr moose said:

 

maybe think about i this way,  the auto manufacturer has 50,000 carby cars in stock, is developing EFI but it's only just better than carby,  they can release now and people will be less interested in their product becasue thins new product isn't much better and the car makers still has 50,000 cars they can't sell.  Or they wait until the 50,000 are almost all sold and put that revenue into making the EFI even better.

 

With the later the consumer gets a better product.

Why does any new product/feature ever get released then? No need to R&D and huge batch sizes would decrease production costs.

Edit: I said sitting on it for 5yrs not selling 50k unit which would be take a month or less.

Just now, mr moose said:

 

You need stable revenue before you have the cash to innovate.  Look at how long AMD was fighting to release half decent CPU.   That's what happens when you don't have steady revenue stream to prop up R+D.   Relying on people buying the latest just because it is is not by itself going to keep the company afloat.

 

The idea is that by maximizing sales gets maximum revenue (even if that means sitting on a product), which means more in the bank to produce a better product tomorrow. AMD failed to maximise sales and that lead to the big lull they had in product development. 

 

 

Pretty much that, but more so if a company loses to much revenue because it release a new product too early, it means less in the bank for R+D which means the consumer gets less the next time around.

Respond to my recent post please-

3 minutes ago, dexT said:

For example: How would it be good if say, the auto manufacturers thought "we have this fuel injection system but we don't want our buyers to have instant gratification so it's carburetors for the next 5yrs"?

6 minutes ago, thorhammerz said:

Good for the consumer: quite debatable.

 

Good for the company: oh very much so.

 

Much of what Mr. Moose has described is simply NVIDIA looking to maximize profits. And lest we forget, a business seeks to serve its own interest (the bottom line), and not the interest(s) consumer base beyond what makes the consumer cough up $$$ to them. 

For example: How would it be good if say, the auto manufacturers thought "we have this fuel injection system but we don't want our buyers to have instant gratification so it's carburetors for the next 5yrs"?

13 minutes ago, mr moose said:

 

I didn't say that was the outcome,  It is the driving factor in demand for product to which manufacturers have to supply.  Cave into the instant gratification demand too soon and fill the coffers versus making them wait and getting products competitive enough to release for a more stable ongoing revenue stream and product advancement. 

 

If you want to see what happens to companies that cash in early on consumer demand look at dell and 3dFX (god rest their soul ), they saw a dollar in making their own boards as there was a consumer demand for right now, they bought STB and that was essentially their undoing.  AMD neasrly went the same way.  Dell became huge by cutting out the middle man in custom PC sales,  They sold and rose quickly, but to their own demise because they hadn't planned beyond that. when none of their customers needed to upgrade and the rest of the industry had caught on (online sales) they are now chasing the pack, not leading it.

 

Every industry is the same,  a balance between moving forward, not stagnating and not fucking up your own market.  Sandbagging is an essential method for controlling supply demand and product value.

No. Innovation will drive sales, look at Ryzen.

They know a large amount of people will buy the next series of cards even if they just bought a 1080/ti/titan so yes, they want to milk the pascal cow. This is stagnation.

Edit: How is delayed gratification good for the consumer? They're sitting on a market ready Volta right now.

2 minutes ago, mr moose said:

Insults are not valid rebuttals.  And against the standards.

Ok how is "Delayed gratification versus instant gratification for the consumer" a positive outcome for the consumer exactly?

Report me.

3 hours ago, asand1 said:

Sorry to hear. What happened?

Super-chilled acetone boiled over when my extension deformed and caused condensation. I think I'm done with dry ice and need to move to LN2, It's nasty af anyways.

2 hours ago, mr moose said:

Delayed gratification versus instant gratification for the consumer.  But for the company balancing their product availability against consumer demand they can maximise the money they have to invest in the next product.  Meaning overall you get better product advances (very necessary for any competing company) All companies do this when they can because it is a financial necessity for survival.

 

So Nvidia could release volta now which would eat into the remaining 10 series sales which would lower their overall revenue thus giving them less to invest and maintain profits. 

Wow. All I can say is

3 minutes ago, mr moose said:

You say that like sandbagging has no positive outcomes for consumers.

 

 

I don't care. I'd rather have cutting edge super fast hardware on the market than these positive outcomes(wat?).

Just now, i_build_nanosuits said:

yeah...you tried they were a bargain back in the day thought...they are the same as the core i7 chips but they have the graphics portion disabled,...but they where like 50$ cheaper.

The 1230/1 v3 were the price of the i5K. About $250 new.

This is sandbagging. Release the f'ing Volta already.

15 minutes ago, Drak3 said:

 

On 10/13/2017 at 12:19 PM, JaMGeR said:

AMG I'm so different from everyone !! I'm the only one that doesn't use leds not any kind of light on my PC. I'm so unique scientist will study my lightless PC and corpse when I die. /S . The hell ppl complain about everything, ppl like leds Let them have some, if you don't like just don't get any and move on. I for example like both ways.. having and RGB glass panel one on the living room and one without any led on my room as I find them annoying to sleep with those in my bedroom. 

RGB stuff is the idiot test. If you're dumb enough to dump hundreds of dollas on RGB components and peripherals instead of getting better CPU/GPU/RAM/MOBO/SSD/PSU or cooling.

To each his own though.

That just looks cheesy. You can get Gentle Typhoons for half of that, WTF Corsair?

1 minute ago, utroz said:

Actually DDR3 should be faster by a bit at the same speeds and timings due to the design of DDR4. This has always been true since going from DDR to DDR2, and DDR2 to DDR3 (and maybe even before like FPM to EDO and SDR to DDR..)

I can see that. My DDR2 1400 C5 tears it up and scores higher in a lot of benches using the same CPU than faster DDR3/higher CL but still being nearly even in the latency mathematical equation.

Should be same perf at same freq/timings.

It's a damn grab bag not gambling lol. A grab bag for idiots.

Haha for a second I pictured the equivalent to an addicted gambler but with lootcrates... sucking dicc or robbing people to buy lootcrates. Losing job and family over lootcrates

I killed my board

On 10/13/2017 at 1:58 PM, GoodBytes said:

Gigabyte? Don't you mean Gigabyte Aorus 1070TiXpKFTW G1 OC edition RGB Fusion Liquid Limited Founders 16GB Revision 2.0 with Gigabyte Ultra Durable 4.0 and SmartFan 5.0 ?

 

You forgot Wind/Waterforce and a few others. So Gigabyte Aorus 1070TiXpKFTW G1 OC edition RGB Fusion Liquid Limited Founders 16GB Revision 2.0 Waterforce Xtreme Gaming with Gigabyte Ultra Durable 4.0 and SmartFan 5.0

6 minutes ago, Ground said:

Well, let's see. I have the E5620 dialed in at 240x19, lets see if it actually runs... The highest I had before was 230x18, so no matter what, it'll be an improvement.

Open task manager processes and kill your realtime virus scanner if you have one. Right click Cinebench in processes and set priority to realtime or high if you can't do RT. Run it four times without closing the R15 window letting it cool fully between runs and use the best one. Screenshot each run and close cpuz before each run.