News PassMark sees the first yearly drop in average CPU performance in its 20 years of benchmark results

[bit_user]

This data is incomplete you can not get an average score from less users. You need the user count to be equal or near equal to accurately say the average score is higher or lower, so this information is incomplete.

The number of submissions only needs to be statistically significant, which is easily achieved at probably just a few hundred submissions. However, that assumes there's no bias in their data, which there surely is. Even having more submissions won't necessarily eliminate bias.

 

[spongiemaster]

This is incorrect. Buying power for the median worker has increased since 2019 to 2024. This is a common misconception since inflation started increasing. We collectively have even more money compared to the cost of goods after all the inflation for a while now. The tariffs will certainly start to cause more inflation again and reduce our buying power in the coming months / years.

"There are three kinds of lies: Lies, Damned Lies, and Statistics"

I have no idea how they are calculating these numbers, but they aren't reality and it isn't difficult to prove that.

Median US household income decreased from 2019 to 2023. Having a higher payrate doesn't mean much if you're working fewer hours or your SO never reemployed after covid.

Median household income U.S. 2023 | Statista

The average salary in the United States has been increasing since 1990. Generally speaking, the median income increases each year.

www.statista.com

Over that same time frame, the average house price in US went up 35%.

https://fred.stlouisfed.org/series/ASPUS

Average new car price in the US didn't exactly go down over that period either. Up by about the same 35% as housing.

The two biggest expenses for most people, housing and transportation, have increased significantly since 2019, While household income dropped from 2019 to 2023. Trying to claim the average person has more disposable income today than 5 years ago is comically false.

 

[bit_user]

I have no idea how they are calculating these numbers, but they aren't reality and it isn't difficult to prove that.

The answer is obvious. Helper's data cites worker pay, whereas yours cites household income. If more households become single-earner, they can both be true.

We know that in the pandemic, women disproportionately dropped out of the labor force. That would hit household income but not worker pay.

 

[dalek1234]

I think that Passmark is sampling only Intel machines, so the graphs make perfect sense 😈🤣

 

[spongiemaster]

The answer is obvious. Helper's data cites worker pay, whereas yours cites household income. If more households become single-earner, they can both be true.

We know that in the pandemic, women disproportionately dropped out of the labor force. That would hit household income but not worker pay.

Doesn't hold water. Median individual income in the US only increased about 16% from 2019 to 2023. Doesn't cover the cost increases I listed above.

https://fred.stlouisfed.org/series/MEPAINUSA646N

Article from the NY Times from 2023 reporting the average American renter is rent burdened for the first time. That's 30% or more of take home pay spent on rent. Rent and vehicle expenses are increasing at a higher rate than salary increases.

https://www.nytimes.com/2023/01/25/realestate/rent-burdened-american-households.html

 

[bit_user]

Doesn't hold water. Median individual income in the US only increased about 16% from 2019 to 2023. Doesn't cover the cost increases I listed above.

Inflation and purchasing power is computed based on a "basket of goods" (i.e. an average over many expenses). You cherry-picked housing and cars, which are two of the things that increased the most.

 

[SpenReyn2]

Average CPU performance across the globe dropped for the first time in over 20 years.

PassMark sees the first yearly drop in average CPU performance in its 20 years of benchmark results : Read more

Could it be that PassMark is the same crappy software it's always been? It was well documented during the zen3 rollout that they tweaked the software to make Intel processors look better. Could it just be that enthusiasts don't use PassMark anymore?

 

[logainofhades]

Could it be that PassMark is the same crappy software it's always been? It was well documented during the zen3 rollout that they tweaked the software to make Intel processors look better. Could it just be that enthusiasts don't use PassMark anymore?

You sure you aren't confusing PassMark with Userbenchmark?

 

[randomizer]

Read it again, it's based on the US and its people;

"We find that in the year ending in the second quarter of 2024, the median American worker could afford the same goods and services as they did in 2019, plus an additional $1,400 to spend or save per year. "

Or are you referring to something other than my post you quoted?

I should have been clearer. I was referring to the passmark graphs.

 

[shady28]

Inflation and purchasing power is computed based on a "basket of goods" (i.e. an average over many expenses). You cherry-picked housing and cars, which are two of the things that increased the most.

Both real (inflation adjusted) median personal (individual) income and household income fell since 2019. See charts below from the Federal Reserve.

This has also been a regional thing. Real income in Florida for example is much higher now, while in New York and California it's worse than the national numbers.

 

[bit_user]

Both real (inflation adjusted) median personal (individual) income and household income fell since 2019. See charts below from the Federal Reserve.

This has also been a regional thing. Real income in Florida for example is much higher now, while in New York and California it's worse than the national numbers.

A few things about that...

First is the use of R-CPI-U-RS, which is described here and includes noted limitations:

https://www.bls.gov/cpi/research-series/r-cpi-u-rs-home.htm
​

Second, it includes transportation and housing. In particular, housing inflation is being exacerbated by the generational bulge of the gen Z now having reached the age where they're piling into the home market. Housing is a fairly inelastic market, so we see substantial price response to changes in demand.

Population pyramid of the United States in 2018
Source: https://en.wikipedia.org/wiki/Generation_Z_in_the_United_States
​

Also, post-pandemic, a lot of people who had been living together now wanted their own place, further overheating the housing market.

What's interesting about the housing component is that existing home owners didn't feel it. Same thing about car prices, if you weren't looking to buy one. So, that's where R-CPI-U-RS can be misleading.

Unfortunately, they don't have inflation-adjusted data that excludes housing and transportation, but this shows what happened to unadjusted median personal income:

​

P.S. FRED is a service provided by the Federal Reserve Bank of St. Louis.

Federal Reserve Economic Data | FRED | St. Louis Fed

Download, graph, and track 825,000 economic time series from 114 sources.

fred.stlouisfed.org