The "halo effect" - is anyone really that dumb?

[swaaye]

Yeah, you know what I'm talking about. Sounds like you've spent a bit of time in the trenches too.

I've been running into a lot of really mindless "puter experts" myself.

I had a neighbor who gave me the most ridiculous explanation of IPv6 and then told me the point isn't at all to extend the number of IPs. This same guy installed XP on a P2 with 64MB RAM as well. That is most disconcerting to me. That someone who has degrees in the IT industry can be a complete idiot. I guess I knew that already but just needed a nice, solid real-world example of the hypothesis.

What's even more amazing is that the same thing happens in areas of fairly notable complexity, like, oh, emergency medicine. Hell it happens in everything.

 

[Entropy]

Of course there is some effect to having the top monster.
The real question is - how big is the effect?

My wife is a graphical designer and has a company that deals a lot with branding. So we have had quite a few scientist vs designer discussions over the breakfast table over the years. And one thing is sure - branding has effect, a specific occurence such as this has smaller effect, and the value is always situational.

The original poster had valid points. Either you don't care, and you buy what's inside/recommended, or you care, and then you are bound to realize that there are different cards, and what you are buying is not the heaviest hitter. Also, we have very recent empirical evidence of ATI having the arguably fastest offering in the X19x0XTX for a reasonable period of time, and still having their marketshare eroding horribly, indicating that if the halo effect is there, it is pretty weak.

That is not unique to graphics cards btw - Mercedes Benz has a good brand, but that doesn't make their compact car sell like hotcakes to people believing they get a bargain.

There are a couple of factors however that could make the shine of having the fastest card strenghten the other products in the line. The most important is probably the suggestion that there is a halo effect. So if nVidia talks with Dell, they will point to the charts and say "we're at the top, consumers see us at the top, and nVidia is the brand they will prefer, so it's in your best interest that you offer our products!". And that might turn out to be a relevant bargaining chip, leading to high volume sales. Note that whether the end user actually buys into this line or not is irrelevant as long as the Dell negotiator feels that the end user might. And I think this thread amply demonstrates that the "I'm smart, but the rest of the world are idiots!" mentality is well spread. And probably applies to Dell/HP/etc product strategists as well.

I have made no secret of the fact that I think the focus on "the fastest" in enthusiast tech publishing is exaggerated and strengthen problematic trends. Yes it interesting, but is it really all that interesting or important? I don't think so. But then again what is the industry going to hype otherwise? Well, as long as the marketeers assume that their customers are pea-brained young males, they could try for even bigger boobies on the packaging, and in fact they do so. Along with Hyper-super-XTX-Ultra-Golden Sample naming, which is also used by all players, indicating what level of customer they assume they address. Hyping the fastest/biggest/noisiest (remember the dustbuster "badge of honour"!) of anything is very very simple marketing....

Neither me nor my wife, (I asked her although her lack of interest was monumental), can see any reasonable way to actually assess the importance of the halo effect, other than noting that what feeble empirical data exists doesn't indicate that it's very strong.

 

[JoshMST]

Oh man, the amount of times I have seen/heard "Dude, GeForce rocks!" "But I can only spend $90!" "Dude, just get the GeForce at that price!"

They probably also have a "Pententium" processor as well...

 

[nicolasb]

Well, as long as the marketeers assume that their customers are pea-brained young males, they could try for even bigger boobies on the packaging, and in fact they do so.

Now, see, that strikes me as an eminently reasonable and practical policy, and one that should encouraged.

 

[Entropy]

Now, see, that strikes me as an eminently reasonable and practical policy, and one that should encouraged.

Heh.
Well, I've got nothing against rendered boobs per se, but when, some time ago, I went shopping for a graphics card for my Sims2 playing 14 year old niece, it really struck me as negative. I felt a bit embarrassed having to queue with a package that screamed of pimply-faced sexuality, and didn't feel too thrilled about giving the box to my niece either. By trying too hard to attract a certain demographic, they ended up turning away another, equally valid, customer.

This was only the packaging, but using a car analogy, it's as if all Toyotas would have flaming skulls painted on them. Not everyone would see that as "cool" either, and it wouldn't really help either sales or public perception.

As far as I can see, the industry has been painting itself further and further into a corner by catering to its most apparent audience. In the context of this thread, by overemphasising the top performers. It generates some buzz (at least when a new top end card is released) but it is also sending the message that PC gaming is very expensive, requires frequent upgrades, and is the domain of the PC/tech enthusiast. Targeting the obvious core demographic too strongly isn't conductive to growth.

Regarding halo effect of top performing cards, I'll reiterate:
1. Of course it exists, but the more interesting and difficult question concerns how strong the effect really is.
2. There are different mechanisms. It probably doesn't affect direct purchases much but may affect indirectly, as per my Dell example above. (The gift industry is an interesting example where all marketing is geared towards creating the impression that a particular item is what somebody else would like to have.)
3. It is a valid question if trying to drum up a halo effect around the top end monster cards isn't actually counter productive when looking at the PC gaming scene as a whole.

Just food for thought.

 

[trinibwoy]

I completely agree with your distaste for some retail packaging but I don't see how that's relevant to the halo-effect. I don't really understand who these people are targeting with some of these images but they certainly don't indicate any particular level of performance. I think it simply has to do with the number of competitors among AIB's and the need to differentiate.

It is a valid question if trying to drum up a halo effect around the top end monster cards isn't actually counter productive when looking at the PC gaming scene as a whole.

Promoting the flagship product is something all companies do in every industry. Exactly why are you trying to find some deeper meaning in this specific to the 3D graphics industry? The halo effect is a simply a passive side effect of the marketing of high-end products due to human nature and isnt driven by anything companies explicitly do.

As far as I can see, the industry has been painting itself further and further into a corner by catering to its most apparent audience. In the context of this thread, by overemphasising the top performers. It generates some buzz (at least when a new top end card is released) but it is also sending the message that PC gaming is very expensive, requires frequent upgrades, and is the domain of the PC/tech enthusiast. Targeting the obvious core demographic too strongly isn't conductive to growth.

That's an interesting point. But the pace of development in the industry is a direct result of competition, advancement in inputs (semiconductor process) and the flexibility of the PC platform. You would have to make a concerted effort not to innovate and that would spell death unless you are the only game in town.

Though, it is kind of ironic that much of this sentiment was born out of ATI's inability to compete at the high-end but they don't seem to be doing too hot at any other price point either.So what's the moral of this whole story?

 

[ban25]

"I'm going to buy the worst possible product available in my price range because the company that makes that product also produces the best possible product in a completely different price range."

That, sir, is a textbook strawman.

 

[poopypoo]

Entropy, I agree with your point! Trini's correct too, but that doesn't mean it's not something people should do something about. It's a little like saying global warming is something that developed naturally with competition -- sure it did, but that doesn't mean it's not counterproductive to ignore it now that we can identify it.

However, the obstacles do seem puzzling -- how to embrace the other half without alienating the core market? -- one thing i can think of is, perhaps marketing of such products towards females and noobs could begin with application-specific marketing? Like a "Sims 2" model graphics card -- Find a card in your lineup that a) you want to sell a lot of b) works well for the Sims (basically any modern card, so pick a decent value model) and c) has any other features that might benefit the "casual" gamer -- quiet, single slot, fits well in a cramped Dell case, good video decoding perhaps -- and then run off a few thousand in a white & blue packaged "Sims 2" bundle or something. If that works, perhaps more extreme maneuvers would, too. The toughest part would have to be that, applicationwise, even the PC market has evolved to a pretty nerdy place as well. Think of all the more gender-unspecific (or female-specific) titles, and most of them (with the glaring exception of the Sims) are Japanese and console-exclusive. I suppose other builder-type games are universally popular... I've known a number of women into Civ and AoE and SimCity. Regardless, the Sims seems the safest choice. That's my two cents!

 

[Entropy]

What the industry can do to change course is probably a thread of its own.

To assess the influence of the high-end cards, I think it would be wise to take a couple of steps back, look at the market share data, and really let the message sink in. The market, as a whole, is pretty much completely uninterested in the high-end. Incidentally this implies (but does not prove) that the halo-effect of these parts aren't much to write home about either. Not only isn't a high-end halo effect strong in other markets, graphics cards are so cheap that it isn't very needed either. People who make their own money can afford to buy the high end, even people who don't make money can often scrunge up sufficient funds. If they wanted. The market data tells it like it is - apart from a very small percentage, people just aren't interested in buying the high end parts. For whatever reasons. They don't fit into their laptops, they are overkill in features (particularly bad right now with the DX10 tie in to Vista), they draw too much power, they are too noisy, they cost too much in relation to the incremental benefit they bring, it is too much money to spend on a toy. Whatever.
The data is crystal clear however in terms of actual market relevance. It is very small. Which may be one reason for people who are interested in this segment to happily buy into halo-effect theories because it lends greater significance to the high end than it manages on its own.

Beyond3D is a place for very 3D interested people to aggregate. The forum biases can be quite strong.

 

[trinibwoy]

Entropy, I may be misreading you but to me all you're saying is that more people buy cheaper products. That isn't exactly a groundbreaking revelation is it. Given the context of this discussion - the marketing influence of high-end products - I'm curious why you assign so much importance to a market segment that is fundamentally indifferent to which GPU ends up in their PC and in some ways oblivious to Nvidia and ATI and the 3D drama between them.

The question you are posing is "How much influence does the high-end have on consumer choice". Yet you are trying to answer it by including those consumers that do not make a choice. Maybe you should change your question to "How many people buy high-end products?" Because that's the question you seem to be addressing.

What percentage of your all encompassing universe of consumers actually makes a conscious choice of which graphics card ends up in their machine? How about you assess the influence of high-end cards among those consumers who actually know that they exist? If your point is that the high-end has little influence because most people aren't even exposed to it then that's a fair argument. But the influence of the high-end in the online community is palpable.

 

[nicolasb]

That, sir, is a textbook strawman.

Why thank you, you're too kind.

 

[Frank]

How about the GeForce 5200? The second best sold card ever, after the GeForce 2/4 MX? People bought it because "nVidia is the best!". And it's cheap.

Why do people buy a Mercedes, BMW or Audi, for twice what they would pay for a better car from another brand? Status.



I was asked to recommend a new PC for a heavy gamer a few weeks ago. I told him, that he really wanted a GeForce 8800 GTX and a C2D 6600E. He complained: "But, isn't there something better from ATi? I like ATi! They're the best! And their top-range card is a great bargain! Why would I pay more for something worse from nVidia? And, I really like AMD for processors! Athlons are much better than Pentiums! AMD rocks!"

 

[overclocked_enthusiasm]

Just ask Steve Jobs about the Ipod and the halo effect on his PC sales. I have made the arguement in another thread that Intel was able to maintain superior pricing to AMD for the last 6-9 months because of Core2Duo. This halo effect even helped them clear some of their Netburst P4 inventory which is inferior to AMD yet people knew that "Intel has the best chips right now" so they bought anything Intel.

The masses are a thundering herd of idiots running to the brightest neon sign with hands outstreched screaming "me too!!". Besides techies on this thread and others like it, people have no idea except what other people tell them. Marketing 101 is about top of mind awareness and brand recognition. Acheive those 2 things and sales will follow every time.

So, "is anyone really that dumb"? Absofrickinglutely.

Consider your knee jerk reaction to the following:

Best video card maker
Best CPU maker
Best thermal paste
Best PSU
Best keyboard/mouse

At some point the companies you thought of had the "best" and the residual predisposition for you to buy them again is what they worked so hard to acheive. Some may have been very recent or some may have been in the past...either way they captured you as a customer. There are some that are immune to this marketing and shop purely on price, value, etc. but they are the exception and not the rule.

 

[poopypoo]

Yeah the best example of this for me is Best Hard Drive company. We've several threads in this very forum where we all disagree on this rather fundamental computer-building issue. For me, I can admit that I haven't even tried Hitachi in one of my own builds but ultimately what matters to me most is this aura of reliability around Seagate (well, and obviously I'm happy with their reliability since I switched to them a few years back). Their warranty and their fairly spotless record with the professional community is worth the $8 USD premium I pay on their drives, and with hard drive pricing being so competitive, I'm sure that makes a huge difference to them. All this, almost entirely on brand recognition. I do check the reviews when before I buy, just in case Hitachi has designed the Millenium Falcon of HDDs last week, but assuming the benchmarks are close enough...

 

[Skrying]

Yeah the best example of this for me is Best Hard Drive company. We've several threads in this very forum where we all disagree on this rather fundamental computer-building issue. For me, I can admit that I haven't even tried Hitachi in one of my own builds but ultimately what matters to me most is this aura of reliability around Seagate (well, and obviously I'm happy with their reliability since I switched to them a few years back). Their warranty and their fairly spotless record with the professional community is worth the $8 USD premium I pay on their drives, and with hard drive pricing being so competitive, I'm sure that makes a huge difference to them. All this, almost entirely on brand recognition. I do check the reviews when before I buy, just in case Hitachi has designed the Millenium Falcon of HDDs last week, but assuming the benchmarks are close enough...

That's actually ironic. Like Seagate made the quietest hard drives for awhile (or so was the thought) be these days thats not really the case, Samsung drives seem to be consistently quieter. Hitachi drives actually do tend to out perform other drives, not by a huge margin but if there is a performance leader it is Hitachi. I tend to stick with a good all around drive, which has lead me to Western Digital a lot ironically, plus they are always cheaper it seems when I go looking.

 

[_xxx_]

Why do people buy a Mercedes, BMW or Audi, for twice what they would pay for a better car from another brand?

Now you're exaggerating quite a bit here! It's neither twice as expensive nor are the half-as-expensive brands better in any possible way. Now if you're saying "cheaper brands have better price/performance ratio", there we may agree.

 

[Entropy]

Entropy, I may be misreading you but to me all you're saying is that more people buy cheaper products. That isn't exactly a groundbreaking revelation is it. Given the context of this discussion - the marketing influence of high-end products - I'm curious why you assign so much importance to a market segment that is fundamentally indifferent to which GPU ends up in their PC and in some ways oblivious to Nvidia and ATI and the 3D drama between them.

Because these are the people who are supposedly affected by the halo of the highest end offerings? That's the definition of a halo effect. And I'm saying that I don't think they are particularly affected by any such halo.
And judging by what you say above, you don't think they are very affected by the top-end performance race either.
(Sorry about taking my time responding. Travelling/jam-packed schedule.)

What percentage of your all encompassing universe of consumers actually makes a conscious choice of which graphics card ends up in their machine? How about you assess the influence of high-end cards among those consumers who actually know that they exist? If your point is that the high-end has little influence because most people aren't even exposed to it then that's a fair argument. But the influence of the high-end in the online community is palpable.

Yes, but is it in any way representative and sales driving? That's why we need to look at hard data, rather than forum opinions. And, as the original poster wrote, the people who actually care enough to keep any kind of track, aren't they also aware that they are not buying the high-end offering and instead try to get a good deal in whatever market segment they are interested in?
The people who make a conscious choice are an interesting group, but difficult or impossible to isolate. How would you do it? But these were the reason I pointed out that ATI lost market share even when having the top performers. They are at least included there.

Just ask Steve Jobs about the Ipod and the halo effect on his PC sales.

Doesn't apply. That alleged effect would be due to people having a hands on positive experience with a given product, leading them to being more positively inclined to try another kind of product from the same manufacturer. Not the case here.

This is more like Toyota manufacturing a super sports car beating all others in 0-60mph, leading people to buy cheaper Toyota models. But the fact is, the automobile market doesn't work that way. (And graphics cards have even less reason to work that way, because the high end offerings are actually financially accessible to a large part of the market.)

I'm not saying the halo effect doesn't exist. I'm just saying that every shred of data I've seen points to it actually being fairly insignificant for top-end graphics cards, particularly as far as the end consumer is concerned.

 

[Entropy]

Yeah the best example of this for me is Best Hard Drive company. We've several threads in this very forum where we all disagree on this rather fundamental computer-building issue. For me, I can admit that I haven't even tried Hitachi in one of my own builds but ultimately what matters to me most is this aura of reliability around Seagate (well, and obviously I'm happy with their reliability since I switched to them a few years back). Their warranty and their fairly spotless record with the professional community is worth the $8 USD premium I pay on their drives, and with hard drive pricing being so competitive, I'm sure that makes a huge difference to them. All this, almost entirely on brand recognition. I do check the reviews when before I buy, just in case Hitachi has designed the Millenium Falcon of HDDs last week, but assuming the benchmarks are close enough...

Ironically, the fastest hard drives have often been made by Fujitsu, they have had excellent 15000rpm offerings. They and Seagate have been competing at the top end for quite a long time.

And it hasn't meant diddly squat to Fujitsus sales in other market segments.
Why would graphics cards be so different?

 

[overclocked_enthusiasm]

If you use gross margins and ASPs of the 2 most recent examples of companies "having the best" and their halo effect on their entire product line, I think you will admit something is there.

Nvidia vs ATI

Intel vs AMD

Do you really think that consumers know the benchmarks between various CPUs or GPUs before they buy? Do they take the time or have the knowledge to even make an informed decision like that? I say that the vast majority of consumers do not and therefore you have to ask yourself what do they rely on to make a decision?

There is anecdotal evidence to show that Intel was able to sell through its inventory of older P4 chips due to the success of Core2Duo. Without the halo effect from Core2Duo why would people buy a P4 over a better AMD Athlon? I submit it is because they either heard "Intel is the best right now" from the salesperson or if they did a little reading of reviews before they bought and heard "Intel is the best". Again, most don't realize the difference between Netburst and Conroe and just bought on the name. Take a look at this http://www.tgdaily.com/images/stories/article_images/processor_performance/20070413/data/data3.jpg chart. Why did AMD have to slash their prices so much more than Intel? Why was Intel not forced to cut P4 prices as much as comparable AMD products? In the end, AMD's gross margins are in the 20's and Intels are in the 50's and Intels' ASPs accross the board are much higher even on older P4 stuff.

I also don't understand why the Ipod can't be used as an example and why bring Toyota into a computer discussion? The halo effect lends clout to other products from the same company due to the success of another product via the resulting goodwill generated for the company itself. Since you brought Toyota into it, how do you think they came to be the #1 carmaker in the world? The Camry was the highest rated, safest and best value for the buck for years and years. Toyota rode to the #1 spot on the back of Camry and allowed them to broaden their product range to now include full sized trucks etc. Without the halo effect from Camry, Toyota would have had a much harder time. Look at Apple's PC market share...it was stuck in the mud. When did it start to move? It was after Ipod took the world by storm.

For a graphics company the halo effect comes generally from the top down unlike the Apple and Toyota examples above. Why? Because the flagship is generally the first card released in a new generation of cards. If this flagship card shows well the goodwill and mindshare generated propel sales of the entire product stack. Since the market for the highest end 7800 GTX or 8800 GTX is so small how do we explain the big jump in Nvidia's gross margins and ASP's compared to ATI during these times? While Nvidia is in the 40% gross margin range ATI is in the 20's. With the midrange Nvidia 88xx cards just being released, why was Nvidia able to outsell ATI and out ASP ATI so badly during the past 6 months? I know R600 and R580 were horribly late but why did consumers buy Nvidia over ATI in general? It was because "Nvidia is the best" and their entire prodcut line sold like hot cakes. In 2 years Nvidia has doubled up ATI in terms of revenue, gross margins and much more than that in profitability when ATI was higher than Nvidia in all these metrics before. It was the 7800 GTX and the 8800 GTX blazing the trail for the entire stack of SKUs from Nvidia. I calll that a pretty impressive Halo effect considering the small percentage of sales these 2 products have on sales compared to the results posted by Nvidia during this time.

Bottom line is the halo effect cuts both ways. ATI and AMD have a negative halo effect and Intel and Nvidia have a positive halo effect. The results of this can clearly be seen in the revenue, gross margins and ASPs of all the companies I mentioned.

 

[Freak'n Big Panda]

I completely agree, the halo effect has a big impact on brand perception which is obviously tied to sales. However I've noticed that the halo effect has a sort of lag time, i.e. consumers don't adjust immediately to a new performance king. A good example would be ATI after the release of the R300, Nvidia Geforce FX boards still sold well during this time period despite far worse performance.