Let us set terms straight first. The resolution is a dimensionless quantity, calculated by nu/Dnu (nu being the absolute frequency, not the Raman shift), therefore 10cm-1 is not a suitable resolution indication.

I assume you are referring to the FWHM which you could use as FWHM=Dnu in the resolution calculation. If you only want to identify species and not make shift analyses, an FWHM of 14cm-1 is sufficient. If you want to perform analyses based on peak shifts, e.g. for stress checks, any accuracy you can gain at an acceptable price is welcome.

The answer is strongly depends from application field you planned to. Spectral resolution means the possibility to resolve two peaks which is close as spectral resolution. Thus, if you need decide is two samples equivalent or not then 14cm-1 for some cases could enougth (for examples ruby mineral and glass). But if you want decide is the ruby mineral natural or syntetic then even 10 cm-1 not enought. The equipment with spectral resolution you choose sutable for quick analysis outdor. There are more inorganic materials for which raman lines FWHM in range 2-6 cm-1, and for analysis you need higer resolution equipment.

As pointed out by Jürgen Weippert you should be careful about you specify your parameters. Although it does not define resolution, what is commonly quoted for (dispersive) Raman spectrometers is the dispersion, in terms of wavenumbers/pixel. I suspect that this is what you are referring to. Then, as indicated by Alexander S Krylov , your requirements depend on the application field. Biological or organic macromolecules tend to have broad Raman features, whereas crystalline inorganic materials have very narrow band widths.

I would also mention that for small molecules as you lower the temperature the line widths narrow. Having more resolution can pull out details. Pay for the higher resolution model if you can. Eventually you will need it as "common" samples turn out not to exist.

This is one of the more often raised questions from customers, the second one is which laser shall I use for my application.

the way to answer these questions is solely based on your application and your expected use of the data. I see that this has been so far the most common response to your question.

if you require to run some basic chemical identification (and you have the library to search), it is not critical to have 8-10-15cm-1 spectral resolution (be aware that spectral resolution refers to the capacity to differentiate 2 peaks), while many people misunderstand the term of (pixel) resolution.

If you are a new user in the Raman world, I will suggest a 785nm laser -as you will have fewer fluorescence problems_ if you are looking for minerals, no need to go too high in spectral range; (2200 is enough); if you want to analyse organics, then, you will need a larger range (3100cm-1)

if you want to start doing more complex measurements, like quantitative analysis, and chemometrics; you need to speak with a sales specialist, who can guide you in a more suitable product for you.

check my website at www.elodiz.com in the app notes and knowledge some extra pieces of info on how to select your Raman device